Search.setIndex({"docnames": ["contributing/code-style", "contributing/doc", "contributing/how-to-create-a-recipe", "contributing/index", "faqs", "huggingface/index", "huggingface/pretrained-models", "huggingface/spaces", "index", "installation/index", "model-export/export-model-state-dict", "model-export/export-ncnn", "model-export/export-onnx", "model-export/export-with-torch-jit-script", "model-export/export-with-torch-jit-trace", "model-export/index", "recipes/Non-streaming-ASR/aishell/conformer_ctc", "recipes/Non-streaming-ASR/aishell/index", "recipes/Non-streaming-ASR/aishell/stateless_transducer", "recipes/Non-streaming-ASR/aishell/tdnn_lstm_ctc", "recipes/Non-streaming-ASR/index", "recipes/Non-streaming-ASR/librispeech/conformer_ctc", "recipes/Non-streaming-ASR/librispeech/index", "recipes/Non-streaming-ASR/librispeech/pruned_transducer_stateless", "recipes/Non-streaming-ASR/librispeech/tdnn_lstm_ctc", "recipes/Non-streaming-ASR/librispeech/zipformer_ctc_blankskip", "recipes/Non-streaming-ASR/librispeech/zipformer_mmi", "recipes/Non-streaming-ASR/timit/index", "recipes/Non-streaming-ASR/timit/tdnn_ligru_ctc", "recipes/Non-streaming-ASR/timit/tdnn_lstm_ctc", "recipes/Non-streaming-ASR/yesno/index", "recipes/Non-streaming-ASR/yesno/tdnn", "recipes/Streaming-ASR/index", "recipes/Streaming-ASR/introduction", "recipes/Streaming-ASR/librispeech/index", "recipes/Streaming-ASR/librispeech/lstm_pruned_stateless_transducer", "recipes/Streaming-ASR/librispeech/pruned_transducer_stateless", "recipes/Streaming-ASR/librispeech/zipformer_transducer", "recipes/index"], "filenames": ["contributing/code-style.rst", "contributing/doc.rst", "contributing/how-to-create-a-recipe.rst", "contributing/index.rst", "faqs.rst", "huggingface/index.rst", "huggingface/pretrained-models.rst", "huggingface/spaces.rst", "index.rst", "installation/index.rst", "model-export/export-model-state-dict.rst", "model-export/export-ncnn.rst", "model-export/export-onnx.rst", "model-export/export-with-torch-jit-script.rst", "model-export/export-with-torch-jit-trace.rst", "model-export/index.rst", "recipes/Non-streaming-ASR/aishell/conformer_ctc.rst", "recipes/Non-streaming-ASR/aishell/index.rst", "recipes/Non-streaming-ASR/aishell/stateless_transducer.rst", "recipes/Non-streaming-ASR/aishell/tdnn_lstm_ctc.rst", "recipes/Non-streaming-ASR/index.rst", "recipes/Non-streaming-ASR/librispeech/conformer_ctc.rst", "recipes/Non-streaming-ASR/librispeech/index.rst", "recipes/Non-streaming-ASR/librispeech/pruned_transducer_stateless.rst", "recipes/Non-streaming-ASR/librispeech/tdnn_lstm_ctc.rst", "recipes/Non-streaming-ASR/librispeech/zipformer_ctc_blankskip.rst", "recipes/Non-streaming-ASR/librispeech/zipformer_mmi.rst", "recipes/Non-streaming-ASR/timit/index.rst", "recipes/Non-streaming-ASR/timit/tdnn_ligru_ctc.rst", "recipes/Non-streaming-ASR/timit/tdnn_lstm_ctc.rst", "recipes/Non-streaming-ASR/yesno/index.rst", "recipes/Non-streaming-ASR/yesno/tdnn.rst", "recipes/Streaming-ASR/index.rst", "recipes/Streaming-ASR/introduction.rst", "recipes/Streaming-ASR/librispeech/index.rst", "recipes/Streaming-ASR/librispeech/lstm_pruned_stateless_transducer.rst", "recipes/Streaming-ASR/librispeech/pruned_transducer_stateless.rst", "recipes/Streaming-ASR/librispeech/zipformer_transducer.rst", "recipes/index.rst"], "titles": ["Follow the code style", "Contributing to Documentation", "How to create a recipe", "Contributing", "Frequently Asked Questions (FAQs)", "Huggingface", "Pre-trained models", "Huggingface spaces", "Icefall", "Installation", "Export model.state_dict()", "Export to ncnn", "Export to ONNX", "Export model with torch.jit.script()", "Export model with torch.jit.trace()", "Model export", "Conformer CTC", "aishell", "Stateless Transducer", "TDNN-LSTM CTC", "Non Streaming ASR", "Conformer CTC", "LibriSpeech", "Pruned transducer statelessX", "TDNN-LSTM-CTC", "Zipformer CTC Blank Skip", "Zipformer MMI", "TIMIT", "TDNN-LiGRU-CTC", "TDNN-LSTM-CTC", "YesNo", "TDNN-CTC", "Streaming ASR", "Introduction", "LibriSpeech", "LSTM Transducer", "Pruned transducer statelessX", "Zipformer Transducer", "Recipes"], "terms": {"we": [0, 1, 2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "us": [0, 1, 2, 4, 5, 7, 8, 9, 11, 15, 16, 17, 18, 19, 21, 24, 28, 29, 31, 33], "tool": [0, 4, 35], "make": [0, 1, 3, 16, 18, 21, 33, 35], "consist": [0, 18, 23, 35, 36, 37], "possibl": [0, 2, 3, 9, 16, 21], "black": 0, "format": [0, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "flake8": 0, "check": [0, 21], "qualiti": [0, 17], "isort": 0, "sort": [0, 9], "import": [0, 4, 36, 37], "The": [0, 1, 2, 4, 7, 9, 10, 11, 16, 17, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "version": [0, 8, 9, 10, 16, 18, 19, 21, 23, 24, 28, 29, 35, 36], "abov": [0, 9, 10, 16, 17, 18, 19, 21, 23, 25, 26, 31, 33, 35, 36, 37], "ar": [0, 1, 3, 9, 10, 12, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37, 38], "22": [0, 21, 28, 29, 31], "3": [0, 4, 8, 10, 19, 23, 24, 25, 26, 31, 35, 36, 37], "0": [0, 1, 4, 8, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "5": [0, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "4": [0, 4, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "10": [0, 4, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "1": [0, 4, 8, 10, 12, 13, 14, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "after": [0, 1, 7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "run": [0, 2, 4, 7, 9, 12, 15, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "command": [0, 1, 4, 9, 10, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "git": [0, 9, 10, 16, 18, 19, 21, 24, 28, 29, 31, 35], "clone": [0, 9, 10, 16, 18, 19, 21, 24, 28, 29, 31, 35], "http": [0, 1, 2, 4, 6, 7, 9, 10, 11, 12, 13, 14, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "github": [0, 2, 6, 9, 10, 11, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "com": [0, 2, 6, 7, 9, 10, 11, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "k2": [0, 2, 6, 7, 8, 10, 11, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "fsa": [0, 2, 6, 7, 9, 10, 11, 12, 13, 14, 16, 18, 21, 23, 25, 26, 35, 36, 37], "icefal": [0, 2, 3, 4, 6, 7, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "cd": [0, 1, 2, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "pip": [0, 1, 4, 9, 18, 35], "instal": [0, 1, 4, 5, 7, 8, 10, 23, 25, 26, 31, 35, 36, 37], "pre": [0, 3, 5, 7, 8, 9], "commit": 0, "whenev": 0, "you": [0, 1, 2, 6, 7, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "automat": [0, 7], "hook": 0, "invok": 0, "fail": [0, 9], "If": [0, 2, 7, 9, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "ani": [0, 9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "your": [0, 1, 2, 5, 7, 8, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "wa": [0, 9, 10, 21, 24], "success": [0, 9], "pleas": [0, 1, 2, 7, 9, 11, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "fix": [0, 4, 9, 21], "issu": [0, 4, 9, 21, 36, 37], "report": [0, 4, 9], "some": [0, 1, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "i": [0, 1, 2, 4, 7, 9, 10, 11, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "e": [0, 2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "modifi": [0, 16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "file": [0, 2, 7, 8, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "place": [0, 9, 10, 18, 21, 24], "so": [0, 7, 8, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "statu": 0, "failur": 0, "see": [0, 1, 7, 9, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "which": [0, 2, 7, 10, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 36, 37], "ha": [0, 2, 8, 11, 12, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 33, 35, 36, 37], "been": [0, 11, 12, 18], "befor": [0, 1, 10, 13, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "further": 0, "chang": [0, 4, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "all": [0, 6, 7, 10, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "again": [0, 31], "should": [0, 2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "succe": 0, "thi": [0, 2, 3, 4, 5, 9, 10, 12, 13, 14, 15, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "time": [0, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "succeed": 0, "want": [0, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "can": [0, 1, 2, 6, 7, 9, 10, 11, 12, 13, 14, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "do": [0, 2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "Or": 0, "without": [0, 5, 7, 16, 21, 35], "your_changed_fil": 0, "py": [0, 2, 4, 9, 12, 13, 14, 15, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "sphinx": 1, "write": [1, 2, 3], "have": [1, 2, 6, 7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "prepar": [1, 3, 10], "environ": [1, 16, 17, 18, 19, 21, 23, 24, 28, 29, 31, 35, 36, 37], "doc": [1, 10], "r": [1, 9, 28, 29], "requir": [1, 9, 36, 37], "txt": [1, 9, 16, 18, 19, 21, 24, 28, 29, 31], "set": [1, 9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "up": [1, 9, 10, 16, 19, 21, 23, 24, 25, 26, 36, 37], "readi": [1, 16, 21], "refer": [1, 2, 9, 10, 11, 13, 14, 16, 18, 19, 21, 23, 24, 25, 28, 29, 31, 33, 36, 37], "restructuredtext": 1, "primer": 1, "familiar": 1, "build": [1, 9, 10, 16, 18, 21, 35], "local": [1, 9, 23, 25, 26, 35, 36, 37], "preview": 1, "what": [1, 2, 9, 18, 33], "look": [1, 2, 6, 9, 16, 18, 19, 21], "like": [1, 2, 7, 9, 16, 18, 19, 21, 23, 25, 26, 31, 33, 35, 36], "publish": [1, 10, 17], "html": [1, 2, 4, 9, 13, 14, 23, 35, 36, 37], "gener": [1, 10, 12, 13, 14, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "view": [1, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "follow": [1, 2, 3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "python3": [1, 4, 9], "m": [1, 18, 23, 25, 26, 28, 29, 35, 36, 37], "server": [1, 7, 9, 35], "It": [1, 2, 5, 9, 11, 12, 13, 14, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "print": [1, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "serv": [1, 23, 25, 26, 35, 36, 37], "port": [1, 23, 25, 26, 35, 36, 37], "8000": [1, 31], "open": [1, 8, 10, 17, 18, 21], "browser": [1, 5, 7, 23, 25, 26, 35, 36, 37], "go": [1, 16, 18, 21, 23, 25, 26, 35, 36, 37], "read": [2, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "code": [2, 3, 4, 8, 16, 21, 23, 24, 28, 29, 31, 36, 37], "style": [2, 3, 8], "adjust": 2, "sytl": 2, "design": 2, "python": [2, 9, 10, 13, 14, 16, 18, 21, 23, 25, 26, 35, 36, 37], "recommend": [2, 9, 16, 18, 19, 21, 23, 36, 37], "test": [2, 8, 10, 11, 12, 16, 18, 19, 21, 24, 25, 28, 29], "valid": [2, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "dataset": [2, 4, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "lhots": [2, 8, 10, 16, 18, 21], "readthedoc": [2, 9], "io": [2, 9, 13, 14, 23, 35, 36, 37], "en": [2, 9], "latest": [2, 7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "index": [2, 9, 13, 14, 35, 36, 37], "yesno": [2, 4, 8, 9, 20, 31, 38], "veri": [2, 3, 18, 28, 29, 31, 36, 37], "good": 2, "exampl": [2, 7, 8, 10, 12, 13, 14, 24, 28, 29, 31], "speech": [2, 7, 8, 9, 11, 12, 17, 18, 31, 38], "pull": [2, 16, 18, 21, 33], "380": [2, 29], "show": [2, 7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "add": [2, 16, 18, 19, 36, 38], "new": [2, 3, 7, 9, 16, 17, 18, 19, 21, 23, 24, 25, 26, 31, 35, 36, 37], "suppos": [2, 36, 37], "would": [2, 9, 10, 21, 24, 36, 37], "name": [2, 10, 16, 18, 23, 25, 26, 36, 37], "foo": [2, 12, 14, 16, 21, 23, 25, 26, 35, 36, 37], "eg": [2, 4, 6, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "mkdir": [2, 16, 18, 19, 21, 24, 28, 29, 31, 35], "p": [2, 9, 18, 28, 29, 35], "asr": [2, 4, 6, 8, 9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "touch": 2, "sh": [2, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "chmod": 2, "x": [2, 33], "simpl": [2, 18], "own": [2, 23, 36, 37], "otherwis": [2, 16, 18, 21, 23, 25, 26, 35, 36, 37], "librispeech": [2, 6, 8, 10, 12, 13, 14, 20, 21, 23, 24, 25, 26, 32, 33, 35, 36, 37, 38], "assum": [2, 9, 10, 16, 18, 19, 21, 23, 24, 28, 29, 31, 35, 36, 37], "fanci": 2, "call": [2, 4], "bar": [2, 12, 14, 16, 21, 23, 25, 26, 35, 36, 37], "organ": 2, "wai": [2, 3, 15, 23, 25, 26, 33, 35, 36, 37], "readm": [2, 16, 18, 19, 21, 24, 28, 29, 31], "md": [2, 6, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "asr_datamodul": [2, 4, 9], "pretrain": [2, 10, 12, 14, 16, 18, 19, 21, 24, 28, 29, 31], "For": [2, 6, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "instanc": [2, 6, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "tdnn": [2, 4, 9, 17, 20, 22, 27, 30], "its": [2, 10, 14, 18, 25], "directori": [2, 8, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "structur": 2, "descript": [2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "contain": [2, 8, 10, 11, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37, 38], "inform": [2, 10, 16, 18, 19, 21, 23, 24, 25, 28, 29, 31, 33, 35, 36, 37], "g": [2, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "wer": [2, 9, 10, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "etc": [2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "provid": [2, 7, 9, 10, 11, 12, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37, 38], "pytorch": [2, 4, 8, 18], "dataload": [2, 9], "take": [2, 10, 23, 31, 36, 37], "input": [2, 10, 16, 18, 19, 21, 24, 28, 29, 31, 33], "checkpoint": [2, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "save": [2, 9, 10, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "dure": [2, 7, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "stage": [2, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "": [2, 9, 10, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "definit": 2, "neural": [2, 16, 21], "network": [2, 16, 18, 21, 23, 25, 26, 35, 36, 37], "script": [2, 8, 9, 14, 15, 16, 18, 19, 21, 24, 28, 29, 31, 35], "infer": [2, 10, 12], "tdnn_lstm_ctc": [2, 19, 24, 29], "conformer_ctc": [2, 16, 21], "get": [2, 7, 9, 16, 18, 19, 21, 23, 24, 25, 26, 31, 35, 36, 37], "feel": [2, 35], "result": [2, 6, 7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "everi": [2, 10, 23, 25, 26, 35, 36, 37], "kept": [2, 23, 36, 37], "self": [2, 11, 33], "toler": 2, "duplic": 2, "among": [2, 9], "differ": [2, 9, 16, 17, 21, 23, 33, 35, 36, 37], "invoc": 2, "help": [2, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "blob": [2, 6, 10, 14, 23, 25, 26, 35, 36, 37], "master": [2, 6, 10, 12, 13, 14, 18, 23, 25, 26, 35, 36, 37], "transform": [2, 16, 21, 35], "conform": [2, 12, 13, 17, 18, 20, 22, 23, 25, 35, 36, 37], "base": [2, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "lstm": [2, 11, 14, 17, 20, 22, 27, 32, 34], "attent": [2, 18, 19, 33, 36, 37], "lm": [2, 9, 18, 23, 24, 28, 29, 31, 36, 37], "rescor": [2, 19, 24, 26, 28, 29, 31], "demonstr": [2, 5, 7, 10], "consid": 2, "colab": 2, "notebook": 2, "welcom": 3, "There": [3, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "mani": [3, 36, 37], "two": [3, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "them": [3, 5, 6, 7, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "To": [3, 7, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "document": [3, 8, 10, 26], "repositori": 3, "recip": [3, 6, 8, 9, 10, 16, 18, 19, 21, 23, 24, 28, 29, 31, 33, 35, 36, 37], "In": [3, 4, 7, 9, 10, 12, 13, 14, 15, 16, 18, 19, 21, 24, 28, 29, 31, 33], "page": [3, 7, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "describ": [3, 5, 10, 12, 13, 14, 15, 16, 18, 19, 21, 23, 24, 28, 29, 36, 37], "how": [3, 5, 7, 8, 9, 15, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "creat": [3, 8, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36], "data": [3, 10, 12, 13, 14, 17], "train": [3, 4, 5, 7, 8, 10, 13, 14, 33], "decod": [3, 4, 7, 12, 14, 15], "model": [3, 5, 7, 8, 9, 11, 33], "section": [4, 5, 9, 10, 12, 13, 14, 15, 16, 21], "collect": [4, 9], "user": 4, "post": 4, "correspond": [4, 6, 7], "solut": 4, "One": 4, "torch": [4, 8, 9, 10, 15, 16, 18, 21], "torchaudio": [4, 8, 33], "cu111": 4, "torchvis": 4, "11": [4, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "f": [4, 9, 28, 29], "download": [4, 7, 8, 17], "org": [4, 9, 17, 18, 23, 35, 36, 37], "whl": [4, 9], "torch_stabl": 4, "throw": 4, "error": [4, 9, 21], "when": [4, 7, 15, 18, 21, 23, 25, 26, 36, 37], "specifi": [4, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "cuda": [4, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "while": [4, 9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "That": [4, 23, 35, 36, 37], "cu11": 4, "therefor": 4, "correct": 4, "log": [4, 9, 24, 28, 29, 31], "traceback": 4, "most": [4, 36, 37], "recent": 4, "last": 4, "line": [4, 9, 23, 36, 37], "14": [4, 9, 10, 12, 13, 16, 21, 23, 24, 25, 28, 35, 36, 37], "from": [4, 5, 7, 9, 10, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "yesnoasrdatamodul": 4, "home": [4, 16, 21], "xxx": [4, 10], "next": [4, 7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "gen": [4, 7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "kaldi": [4, 7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "34": [4, 9], "datamodul": 4, "__init__": [4, 10, 16, 18, 21], "23": [4, 9, 16, 18, 19, 21, 28, 29, 31], "util": [4, 9, 21], "add_eo": 4, "add_so": 4, "get_text": 4, "39": [4, 9, 18, 21, 24, 28], "tensorboard": [4, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "summarywrit": 4, "miniconda3": 4, "env": 4, "yyi": 4, "lib": [4, 9], "8": [4, 9, 10, 16, 18, 21, 23, 24, 25, 26, 31, 35, 36, 37], "site": [4, 9], "packag": [4, 9], "loosevers": 4, "uninstal": 4, "setuptool": [4, 9], "58": [4, 21], "find": [5, 6, 7, 9, 10, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "also": [5, 6, 9, 10, 11, 12, 14, 16, 18, 19, 21, 23, 25, 26, 31, 33, 35, 36, 37], "try": [5, 7, 23, 25, 26, 35, 36, 37], "within": [5, 7], "anyth": [5, 7], "space": [5, 8], "youtub": [5, 8, 21, 23, 24, 25, 26, 35, 36, 37], "video": [5, 8, 21, 23, 24, 25, 26, 35, 36, 37], "upload": [6, 7, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "huggingfac": [6, 8, 10, 16, 18, 19, 21, 24, 25, 26, 28, 29, 31, 35], "co": [6, 7, 10, 16, 17, 18, 19, 21, 24, 25, 26, 28, 29, 31, 35], "visit": [6, 7, 23, 25, 26, 35, 36, 37], "link": [6, 9, 10, 11, 23, 25, 26, 35, 36, 37], "search": [6, 7], "specif": [6, 18], "aishel": [6, 8, 16, 18, 19, 20, 38], "gigaspeech": [6, 13, 35], "wenetspeech": [6, 13], "integr": 7, "framework": [7, 12, 23, 36], "sherpa": [7, 11, 12, 13, 14, 35], "need": [7, 9, 10, 11, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "window": [7, 11, 12, 35], "maco": [7, 11, 12, 35], "linux": [7, 11, 12, 35], "even": [7, 9], "ipad": 7, "phone": 7, "start": [7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "address": [7, 10, 18, 23, 26, 35, 36, 37], "recognit": [7, 8, 11, 12, 17, 18, 31, 38], "screenshot": [7, 16, 18, 19, 21, 23, 31, 35, 36], "select": [7, 23, 24, 28, 29, 31, 35, 36, 37], "languag": [7, 16, 18, 19], "current": [7, 9, 18, 33, 35, 36, 37, 38], "chines": [7, 17, 18], "english": [7, 31, 35], "target": 7, "method": [7, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "greedi": 7, "modified_beam_search": [7, 18, 23, 25, 35, 36, 37], "choos": [7, 9, 23, 25, 26, 35, 36, 37], "number": [7, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "activ": 7, "path": [7, 10, 12, 14, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "either": [7, 16, 18, 19, 21, 36, 37], "record": [7, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "click": [7, 9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "button": 7, "submit": 7, "wait": 7, "moment": 7, "an": [7, 9, 10, 12, 13, 14, 16, 17, 18, 21, 23, 26, 31, 35, 36, 37], "bottom": [7, 23, 25, 26, 35, 36, 37], "part": [7, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "tabl": 7, "one": [7, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "subscrib": [7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "channel": [7, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "nadira": [7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "povei": [7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "www": [7, 9, 17, 21, 23, 24, 25, 26, 35, 36, 37], "uc_vaumpkminz1pnkfxan9mw": [7, 9, 21, 23, 24, 25, 26, 35, 36, 37], "2": [8, 10, 12, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "frequent": 8, "ask": 8, "question": 8, "faq": 8, "oserror": 8, "libtorch_hip": 8, "cannot": 8, "share": [8, 9], "object": [8, 9, 16, 18, 19, 23, 31, 35, 36], "attributeerror": 8, "modul": [8, 9, 25, 36], "distutil": 8, "attribut": [8, 21], "export": [8, 9, 16, 18, 19, 21, 24, 28, 29, 31], "state_dict": [8, 15, 16, 18, 19, 21, 24, 28, 29, 31], "jit": [8, 15, 21], "trace": [8, 13, 15], "onnx": [8, 10, 15], "ncnn": [8, 15], "non": [8, 12, 21, 33, 36, 38], "stream": [8, 12, 16, 21, 28, 29, 35, 38], "timit": [8, 20, 28, 29, 38], "introduct": [8, 32, 38], "contribut": 8, "depend": [9, 16, 21, 35], "step": [9, 10, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "order": [9, 16, 19, 21, 24, 28, 29], "matter": 9, "least": 9, "v1": [9, 16, 19, 21, 24, 28, 29], "9": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 31, 35, 36, 37], "alreadi": [9, 10], "don": [9, 13, 16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "t": [9, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "replac": 9, "compil": [9, 16, 18, 21], "against": 9, "strongli": 9, "variabl": [9, 16, 19, 21, 23, 25, 26, 35, 36, 37], "pythonpath": [9, 35], "point": [9, 10, 16, 19, 21, 23, 25, 26, 35, 36, 37], "folder": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "tmp": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "setup": [9, 16, 18, 19, 21, 23, 24, 28, 29, 31, 35, 36, 37], "put": [9, 25, 36], "sever": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "same": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "switch": [9, 16, 21, 26], "just": [9, 33], "about": [9, 18, 23, 26, 35, 36, 37], "virtualenv": 9, "cpython3": 9, "6": [9, 16, 18, 21, 23, 24, 28, 29, 35], "final": [9, 10, 21, 24], "64": [9, 10, 18, 36], "1540m": 9, "creator": 9, "cpython3posix": 9, "dest": 9, "ceph": [9, 10, 16, 18, 21], "fj": [9, 10, 18, 21], "fangjun": [9, 10, 18, 21], "clear": 9, "fals": [9, 10, 16, 18, 21], "no_vcs_ignor": 9, "global": 9, "seeder": 9, "fromappdata": 9, "bundl": 9, "wheel": [9, 35], "via": [9, 13, 14], "copi": [9, 33], "app_data_dir": 9, "root": 9, "v": [9, 21, 28, 29], "irtualenv": 9, "ad": [9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "seed": 9, "21": [9, 10, 16, 18, 21, 28, 29], "57": [9, 21, 24], "36": [9, 18, 21], "bashactiv": 9, "cshellactiv": 9, "fishactiv": 9, "powershellactiv": 9, "pythonactiv": 9, "xonshactiv": 9, "sourc": [9, 10, 16, 17, 18, 21], "bin": [9, 16, 21, 35], "dev20210822": 9, "cpu": [9, 10, 13, 16, 23, 25, 26, 31, 36, 37], "torch1": 9, "nightli": 9, "2bcpu": 9, "cp38": 9, "linux_x86_64": 9, "mb": 9, "________________________________": 9, "185": [9, 16, 21, 31], "kb": [9, 28, 29], "graphviz": 9, "17": [9, 10, 16, 21, 28, 29, 35], "py3": 9, "none": [9, 16, 21], "18": [9, 16, 18, 19, 21, 23, 24, 28, 29, 35, 36, 37], "cach": 9, "manylinux1_x86_64": 9, "831": [9, 18, 29], "type": [9, 10, 16, 18, 21, 23, 25, 26, 31, 33, 35, 36, 37], "extens": 9, "typing_extens": 9, "26": [9, 18, 21, 29], "successfulli": 9, "probabl": [9, 18, 23, 25, 35, 36, 37], "req": 9, "7b1b76ge": 9, "q": 9, "audioread": 9, "soundfil": 9, "post1": 9, "py2": 9, "7": [9, 10, 16, 19, 21, 23, 24, 28, 29, 35, 36], "97": [9, 16], "cytoolz": 9, "manylinux_2_17_x86_64": 9, "manylinux2014_x86_64": 9, "dataclass": 9, "h5py": 9, "manylinux_2_12_x86_64": 9, "manylinux2010_x86_64": 9, "684": [9, 16, 31], "intervaltre": 9, "lilcom": 9, "numpi": 9, "15": [9, 10, 18, 19, 21, 28, 31], "40": [9, 19, 21, 24, 28, 29], "pyyaml": 9, "662": 9, "tqdm": 9, "62": [9, 21, 24], "76": [9, 31], "73": 9, "satisfi": 9, "dev": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "2a1410b": 9, "clean": [9, 16, 18, 21, 23, 24, 25, 26, 35, 36, 37], "toolz": 9, "55": [9, 19, 21, 28], "sortedcontain": 9, "29": [9, 16, 18, 19, 21, 24, 28, 29], "cffi": 9, "411": [9, 21], "pycpars": 9, "20": [9, 10, 16, 18, 19, 21, 23, 24, 28, 29, 31, 36], "112": 9, "pypars": 9, "67": 9, "done": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "filenam": [9, 12, 13, 14, 25, 26, 35, 37], "dev_2a1410b_clean": 9, "size": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "342242": 9, "sha256": 9, "f683444afa4dc0881133206b4646a": 9, "9d0f774224cc84000f55d0a67f6e4a37997": 9, "store": [9, 21], "ephem": 9, "ftu0qysz": 9, "7f": 9, "7a": 9, "8e": 9, "a0bf241336e2e3cb573e1e21e5600952d49f5162454f2e612f": 9, "warn": 9, "built": 9, "invalid": [9, 21], "metadata": [9, 28, 29], "mandat": 9, "pep": 9, "440": 9, "packa": 9, "ging": 9, "deprec": [9, 18], "legaci": 9, "becaus": 9, "could": [9, 16, 19], "A": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 35, 36, 37], "discuss": 9, "regard": 9, "pypa": 9, "sue": 9, "8368": 9, "inter": 9, "valtre": 9, "sor": 9, "tedcontain": 9, "remot": 9, "enumer": 9, "500": [9, 10, 18, 21, 26, 35], "count": 9, "100": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "compress": 9, "308": [9, 16, 18, 19], "total": [9, 16, 18, 19, 21, 23, 24, 31, 35, 36], "delta": 9, "263": 9, "reus": 9, "307": 9, "102": [9, 16], "pack": [9, 36, 37], "receiv": 9, "172": 9, "49": [9, 21, 29, 31], "kib": 9, "385": 9, "00": [9, 16, 18, 19, 21, 24, 28, 29, 31], "resolv": 9, "kaldilm": 9, "tar": 9, "gz": 9, "48": [9, 16, 18], "574": 9, "kaldialign": 9, "sentencepiec": [9, 21], "96": 9, "41": [9, 16, 18, 28, 31], "absl": 9, "absl_pi": 9, "13": [9, 10, 18, 19, 21, 24, 25, 28], "132": 9, "googl": [9, 23, 25, 26, 35, 36, 37], "auth": 9, "oauthlib": 9, "google_auth_oauthlib": 9, "grpcio": 9, "24": [9, 19, 24, 28, 29, 31], "ment": 9, "12": [9, 10, 16, 18, 19, 21, 23, 25, 26, 28, 31, 35, 36, 37], "requi": 9, "rement": 9, "protobuf": 9, "manylinux_2_5_x86_64": 9, "werkzeug": 9, "288": 9, "tensorboard_data_serv": 9, "google_auth": 9, "35": [9, 10, 18, 21, 35], "152": 9, "request": [9, 33], "plugin": 9, "wit": 9, "tensorboard_plugin_wit": 9, "781": 9, "markdown": 9, "six": 9, "16": [9, 10, 14, 16, 18, 19, 21, 23, 24, 28, 29, 31, 35, 36, 37], "cachetool": 9, "rsa": 9, "pyasn1": 9, "pyasn1_modul": 9, "155": 9, "requests_oauthlib": 9, "77": [9, 21], "urllib3": 9, "27": [9, 16, 18, 24, 29], "138": [9, 16, 18], "certifi": 9, "2017": 9, "2021": [9, 16, 19, 21, 24, 28, 29, 31], "30": [9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "145": 9, "charset": 9, "normal": [9, 24, 28, 29, 31, 36], "charset_norm": 9, "idna": 9, "59": [9, 19, 21], "146": 9, "897233": 9, "eccb906cafcd45bf9a7e1a1718e4534254bfb": 9, "f4c0d0cbc66eee6c88d68a63862": 9, "85": 9, "7d": 9, "63": [9, 18], "f2dd586369b8797cb36d213bf3a84a789eeb92db93d2e723c9": 9, "etool": 9, "oaut": 9, "hlib": 9, "let": [9, 16, 21, 35], "u": [9, 16, 18, 19, 21, 31, 35], "08": [9, 21, 24, 26, 28, 29, 31, 35], "19": [9, 10, 16, 21, 24, 28, 29], "main": [9, 16, 21, 33], "dl_dir": [9, 16, 19, 21, 23, 25, 26, 35, 36, 37], "waves_yesno": 9, "49mb": 9, "03": [9, 10, 18, 21, 28, 29, 35], "39mb": 9, "manifest": 9, "31": [9, 21], "42": [9, 16, 21, 31], "comput": [9, 10, 16, 18, 19, 23, 24, 26, 28, 29, 31, 35, 36, 37], "fbank": [9, 10, 16, 18, 19, 21, 24, 28, 29, 31], "32": [9, 16, 18, 19, 37], "803": 9, "info": [9, 10, 16, 18, 19, 21, 24, 28, 29, 31], "compute_fbank_yesno": 9, "52": [9, 16, 21], "process": [9, 10, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "extract": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "featur": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "_______________________________________________________________": 9, "90": 9, "01": [9, 18, 19, 21], "80": [9, 10, 16, 18, 21], "57it": 9, "085": 9, "______________________________________________________________": 9, "248": [9, 18], "21it": 9, "lang": [9, 10, 18, 21, 26], "fcordre9": 9, "kaldilm_6899d26f2d684ad48f21025950cd2866": 9, "csrc": [9, 21], "arpa_file_pars": 9, "cc": 9, "void": 9, "arpafilepars": 9, "rea": 9, "d": [9, 28, 29], "std": 9, "istream": 9, "79": 9, "140": [9, 19], "gram": [9, 16, 18, 19, 23, 24, 26, 28, 29, 36, 37], "89": [9, 16], "hlg": [9, 24, 28, 29, 31], "928": 9, "compile_hlg": 9, "120": 9, "lang_phon": [9, 19, 24, 28, 29, 31], "929": [9, 18], "lexicon": [9, 16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "116": 9, "convert": [9, 21, 35], "l": [9, 18, 28, 29, 31], "pt": [9, 10, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "linv": [9, 18, 21, 31], "931": 9, "ctc_topo": 9, "max_token_id": 9, "932": 9, "load": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "fst": [9, 18, 31], "intersect": [9, 23, 36, 37], "933": 9, "lg": [9, 23, 26, 36, 37], "shape": 9, "66": 9, "connect": [9, 10, 21, 23, 24, 35, 36, 37], "68": [9, 21], "70": 9, "class": [9, 21], "tensor": [9, 16, 18, 19, 21, 23, 31, 35, 36], "71": [9, 21, 24], "determin": 9, "934": 9, "74": [9, 10], "_k2": 9, "raggedint": 9, "remov": [9, 16, 18, 19, 21, 24, 28, 29], "disambigu": 9, "symbol": [9, 18, 23, 36, 37], "87": 9, "remove_epsilon": 9, "935": 9, "92": [9, 21], "arc": 9, "95": [9, 17], "compos": 9, "h": 9, "105": [9, 21], "936": 9, "107": [9, 24], "123": 9, "now": [9, 16, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "cuda_visible_devic": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "gpu": [9, 16, 18, 19, 21, 23, 25, 26, 28, 29, 31, 35, 36, 37], "avail": [9, 10, 16, 18, 21, 24, 28, 29, 31, 35], "case": [9, 10, 23, 25, 26, 35, 36, 37], "segment": 9, "fault": 9, "core": 9, "dump": 9, "protocol_buffers_python_implement": 9, "more": [9, 16, 21, 31, 33, 35, 36], "674": 9, "interest": [9, 23, 25, 26, 35, 36, 37], "given": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 36, 37], "below": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36], "072": 9, "465": 9, "466": 9, "exp_dir": [9, 18, 21, 23, 25, 26, 36, 37], "posixpath": [9, 18, 21], "exp": [9, 10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "lang_dir": [9, 18, 21], "lr": [9, 18, 35], "feature_dim": [9, 10, 16, 18, 21, 31], "weight_decai": 9, "1e": 9, "06": [9, 10, 19, 21, 24, 31], "start_epoch": 9, "best_train_loss": [9, 10], "inf": [9, 10], "best_valid_loss": [9, 10], "best_train_epoch": [9, 10], "best_valid_epoch": [9, 10], "batch_idx_train": [9, 10], "log_interv": [9, 10], "valid_interv": [9, 10], "beam_siz": [9, 10, 18], "reduct": [9, 25], "sum": 9, "use_doub": 9, "le_scor": 9, "true": [9, 10, 16, 18, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "world_siz": 9, "master_port": 9, "12354": 9, "num_epoch": 9, "feature_dir": [9, 21], "max_dur": [9, 21], "bucketing_sampl": [9, 21], "num_bucket": [9, 21], "concatenate_cut": [9, 21], "duration_factor": [9, 21], "gap": [9, 21], "on_the_fly_feat": [9, 21], "shuffl": [9, 21], "return_cut": [9, 21], "num_work": [9, 21], "074": 9, "113": [9, 18, 21], "098": [9, 24], "cut": [9, 21], "240": [9, 16, 31], "149": [9, 21], "200": [9, 10, 16, 21, 28, 29, 31], "singlecutsampl": 9, "206": [9, 21], "219": [9, 18, 21], "246": [9, 18, 21, 28, 29], "357": 9, "416": 9, "epoch": [9, 10, 12, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "batch": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "avg": [9, 10, 12, 13, 14, 18, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "loss": [9, 16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "0789": 9, "848": 9, "5356": 9, "7556": 9, "301": [9, 10, 21], "432": [9, 21], "9972": 9, "best": [9, 16, 19, 21], "805": 9, "2436": 9, "5717": 9, "33": [9, 16, 17, 18, 21, 28], "109": [9, 16, 21], "4167": 9, "121": [9, 24], "325": 9, "2214": 9, "798": [9, 18], "0781": 9, "1343": 9, "065": 9, "0859": 9, "556": 9, "0421": 9, "0975": 9, "810": 9, "0431": 9, "824": 9, "657": 9, "0109": 9, "984": [9, 21], "0093": 9, "0096": 9, "50": [9, 10, 21, 23, 28, 35, 36, 37], "239": [9, 18], "0104": 9, "0101": 9, "569": 9, "0092": 9, "819": [9, 28], "835": 9, "51": [9, 16, 21, 31], "024": 9, "0105": 9, "317": 9, "0099": 9, "0097": 9, "552": 9, "0108": 9, "869": 9, "0102": 9, "126": [9, 21], "128": [9, 21], "537": [9, 21], "192": [9, 21], "249": 9, "250": [9, 18, 24], "lm_dir": [9, 21], "search_beam": [9, 16, 21, 31], "output_beam": [9, 16, 21, 31], "min_active_st": [9, 16, 21, 31], "max_active_st": [9, 16, 21, 31], "10000": [9, 16, 21, 31], "use_double_scor": [9, 16, 21, 31], "193": 9, "213": [9, 31], "259": [9, 16], "devic": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 36, 37], "217": [9, 16, 21], "279": [9, 21], "averag": [9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "userwarn": [9, 18], "floor_divid": 9, "futur": [9, 18, 38], "round": [9, 18], "toward": [9, 18], "trunc": [9, 18], "function": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "NOT": [9, 16, 18, 21, 31], "floor": [9, 18], "incorrect": [9, 18], "neg": [9, 18], "valu": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "keep": [9, 18, 23, 36, 37], "behavior": [9, 18], "div": [9, 18], "b": [9, 18, 21, 28, 29], "rounding_mod": [9, 18], "actual": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "divis": [9, 18], "trigger": 9, "intern": 9, "aten": 9, "src": [9, 35], "nativ": 9, "binaryop": 9, "cpp": [9, 13], "450": [9, 16, 18, 19], "k": [9, 23, 28, 29, 35, 36, 37], "n": [9, 16, 23, 25, 26, 28, 29, 35, 36, 37], "220": [9, 18, 19, 21], "409": 9, "190": [9, 24], "until": [9, 21, 25], "571": [9, 21], "228": [9, 21], "transcript": [9, 16, 17, 18, 19, 21, 23, 24, 28, 29, 35, 36, 37], "recog": [9, 18, 21], "test_set": [9, 31], "572": 9, "ins": [9, 21, 31], "del": [9, 21, 31], "sub": [9, 21, 31], "573": 9, "236": 9, "wrote": [9, 21], "detail": [9, 11, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "stat": [9, 21], "err": [9, 18, 21], "299": 9, "congratul": [9, 16, 19, 21, 24, 28, 29, 31], "first": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "fun": 9, "debug": 9, "variou": [9, 15, 38], "problem": 9, "mai": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37, 38], "encount": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "period": 10, "disk": 10, "optim": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "other": [10, 18, 21, 23, 24, 28, 29, 31, 33, 36, 37, 38], "relat": [10, 16, 18, 21, 24, 28, 29, 31], "resum": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "howev": 10, "onli": [10, 12, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37, 38], "strip": 10, "except": 10, "reduc": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "each": [10, 16, 18, 19, 21, 23, 25, 26, 33, 35, 36, 37], "well": [10, 31, 38], "usag": [10, 13, 14, 24, 28, 29, 31], "pruned_transducer_stateless3": [10, 12, 13, 33], "almost": [10, 23, 33, 36, 37], "dir": [10, 12, 13, 14, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "bpe": [10, 12, 13, 14, 21, 23, 25, 26, 35, 36, 37], "lang_bpe_500": [10, 12, 13, 14, 21, 23, 25, 26, 35, 36, 37], "dict": 10, "host": 10, "csukuangfj": [10, 16, 18, 19, 21, 24, 28, 29, 31, 35], "prune": [10, 18, 20, 22, 32, 33, 34, 35, 37], "transduc": [10, 11, 12, 17, 20, 22, 32, 33, 34], "stateless3": 10, "2022": [10, 18, 23, 25, 26, 35, 36], "05": [10, 16, 18, 19, 21, 29], "lf": [10, 16, 18, 19, 21, 24, 26, 28, 29, 31], "repo": 10, "prefix": 10, "those": 10, "wave": [10, 12, 16, 21], "iter": [10, 14, 23, 25, 26, 35, 36, 37], "1224000": 10, "greedy_search": [10, 18, 23, 25, 35, 36, 37], "test_wav": [10, 16, 18, 19, 21, 24, 28, 29, 31], "1089": [10, 21, 24], "134686": [10, 21, 24], "0001": [10, 21, 24], "wav": [10, 12, 14, 16, 18, 19, 21, 23, 25, 26, 28, 29, 31, 35, 36, 37], "1221": [10, 21, 24], "135766": [10, 21, 24], "0002": [10, 21, 24], "multipl": [10, 16, 18, 19, 21, 24, 28, 29, 31], "sound": [10, 14, 16, 18, 19, 21, 24, 28, 29, 31], "Its": [10, 21], "output": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "09": [10, 16, 18, 19, 21, 35], "02": [10, 18, 21, 23, 29, 35, 36], "233": 10, "265": 10, "reset_interv": 10, "3000": 10, "subsampling_factor": [10, 16, 18, 21], "encoder_dim": 10, "512": [10, 16, 18, 21], "nhead": [10, 16, 18, 21, 23, 36], "dim_feedforward": [10, 18], "2048": [10, 18], "num_encoder_lay": [10, 18], "decoder_dim": 10, "joiner_dim": 10, "model_warm_step": 10, "env_info": [10, 16, 18, 21], "releas": [10, 16, 18, 21, 35], "sha1": [10, 16, 18, 21], "4810e00d8738f1a21278b0156a42ff396a2d40ac": 10, "date": [10, 16, 18, 21], "fri": 10, "oct": [10, 21], "miss": [10, 18, 21], "cu102": 10, "branch": [10, 16, 18, 21, 25], "1013": 10, "c39cba5": 10, "dirti": [10, 16, 21], "thu": [10, 18, 21, 24], "jsonl": 10, "hostnam": [10, 18], "de": [10, 18], "74279": [10, 18], "0324160024": 10, "65bfd8b584": 10, "jjlbn": 10, "ip": [10, 18], "177": [10, 18, 19, 21], "203": [10, 21], "bpe_model": [10, 21], "sound_fil": [10, 16, 18, 21, 31], "sample_r": [10, 16, 18, 21, 31], "16000": [10, 16, 18, 19, 21, 24, 25, 28, 29], "beam": [10, 35], "max_context": 10, "max_stat": 10, "context_s": [10, 18], "max_sym_per_fram": [10, 18], "simulate_stream": 10, "decode_chunk_s": 10, "left_context": 10, "dynamic_chunk_train": 10, "causal_convolut": 10, "short_chunk_s": [10, 36, 37], "25": [10, 16, 21, 23, 28, 29, 31, 36], "num_left_chunk": 10, "blank_id": [10, 18], "unk_id": 10, "vocab_s": [10, 18], "271": 10, "273": [10, 18], "612": 10, "458": 10, "disabl": 10, "giga": [10, 35], "623": 10, "277": 10, "paramet": [10, 13, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "78648040": 10, "951": [10, 21], "285": [10, 18, 21], "construct": [10, 16, 18, 19, 21, 24, 28, 29, 31], "952": 10, "295": [10, 16, 18, 19, 21], "957": 10, "700": 10, "329": [10, 21], "912": 10, "388": 10, "earli": [10, 21, 24], "nightfal": [10, 21, 24], "THE": [10, 21, 24], "yellow": [10, 21, 24], "lamp": [10, 21, 24], "light": [10, 21, 24], "here": [10, 16, 18, 19, 21, 24, 33, 35, 36], "AND": [10, 21, 24], "THERE": [10, 21, 24], "squalid": [10, 21, 24], "quarter": [10, 21, 24], "OF": [10, 21, 24], "brothel": [10, 21, 24], "god": [10, 21, 24], "AS": [10, 21, 24], "direct": [10, 21, 24], "consequ": [10, 21, 24], "sin": [10, 21, 24], "man": [10, 21, 24], "punish": [10, 21, 24], "had": [10, 21, 24], "her": [10, 21, 24], "love": [10, 21, 24], "child": [10, 21, 24], "whose": [10, 18, 21, 24], "ON": [10, 21, 24, 35], "THAT": [10, 21, 24], "dishonor": [10, 21, 24], "bosom": [10, 21, 24], "TO": [10, 21, 24], "parent": [10, 21, 24], "forev": [10, 21, 24], "WITH": [10, 21, 24], "race": [10, 21, 24], "descent": [10, 21, 24], "mortal": [10, 21, 24], "BE": [10, 21, 24], "bless": [10, 21, 24], "soul": [10, 21, 24], "IN": [10, 21, 24], "heaven": [10, 21, 24], "yet": [10, 21, 24], "THESE": [10, 21, 24], "thought": [10, 21, 24], "affect": [10, 21, 24], "hester": [10, 21, 24], "prynn": [10, 21, 24], "less": [10, 21, 24, 31, 36, 37], "hope": [10, 17, 21, 24], "than": [10, 16, 18, 19, 21, 23, 24, 25, 26, 31, 35, 36, 37], "apprehens": [10, 21, 24], "390": 10, "alwai": 10, "note": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "down": [10, 16, 21, 23, 25, 26, 35, 36, 37], "reproduc": [10, 21], "ln": [10, 16, 21, 23, 25, 26, 35, 36, 37], "9999": [10, 25, 26, 35], "symlink": 10, "pass": [10, 16, 18, 19, 21, 23, 25, 26, 33, 35, 36, 37], "max": [10, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "durat": [10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "600": [10, 21, 23, 25, 35, 36, 37], "reason": [10, 36], "support": [11, 12, 16, 18, 21, 23, 25, 26, 33, 35, 36, 37], "perform": [11, 12, 18, 36], "raspberri": [11, 35], "pi": [11, 35], "project": 11, "static": [11, 35], "produc": [11, 23, 25, 26, 35, 36, 37], "binari": [11, 16, 18, 19, 21, 23, 31, 35, 36], "everyth": 11, "tree": [12, 13, 14, 16, 18, 19, 21, 24, 28, 29, 31, 35], "insid": [12, 14], "encod": [12, 14, 16, 18, 19, 21, 23, 24, 25, 31, 33, 35, 36, 37], "joiner": [12, 14, 18, 23, 35, 36, 37], "joiner_encoder_proj": 12, "joiner_decoder_proj": 12, "onnx_pretrain": 12, "proj": 12, "baz": [12, 14], "onnxruntim": 12, "our": [13, 14, 21, 23, 33, 36, 37], "torchscript": [13, 14, 35], "cpu_jit": [13, 16, 21, 23, 25, 26, 36, 37], "confus": 13, "move": [13, 23, 25, 26, 36, 37], "why": 13, "streaming_asr": [13, 14, 35, 36, 37], "emform": 13, "conv_emform": 13, "offline_asr": [13, 23], "lstm_transducer_stateless2": [14, 35], "468000": [14, 35], "three": [14, 16, 18, 33], "encoder_jit_trac": [14, 35, 37], "decoder_jit_trac": [14, 35, 37], "joiner_jit_trac": [14, 35, 37], "jit_pretrain": [14, 25, 26, 35], "tutori": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "learn": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "singl": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "1best": [16, 19, 21, 24, 25, 26, 28, 29], "handl": [16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "automag": [16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "stop": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "control": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "By": [16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "default": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "execut": [16, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "mean": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 33, 35, 36, 37], "musan": [16, 19, 21, 23, 25, 26, 35, 36, 37], "sai": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "thei": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "won": [16, 19, 21, 23, 25, 26, 35, 36, 37], "re": [16, 19, 21, 23, 25, 26, 35, 36, 37], "intal": [16, 19], "initi": [16, 19], "sudo": [16, 19], "apt": [16, 19], "permiss": [16, 19], "commandlin": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "quit": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "often": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "experi": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "num": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "state": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "world": [16, 18, 19, 21, 23, 24, 25, 26, 35, 36, 37], "multi": [16, 18, 19, 21, 23, 25, 26, 33, 35, 36, 37], "machin": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "ddp": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "implement": [16, 18, 19, 21, 23, 25, 26, 33, 35, 36, 37], "present": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "later": [16, 19, 21, 23, 24, 25, 26, 28, 29, 35, 36, 37], "second": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "over": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "utter": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "pad": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "oom": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "v100": [16, 18, 19, 21], "nvidia": [16, 18, 19, 21], "due": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "usual": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "larger": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "caus": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "smaller": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "increas": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "tune": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "weight": [16, 19, 21, 25, 26, 35], "decai": [16, 19, 21, 25, 26, 35], "warmup": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "get_param": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "realli": [16, 19, 21, 23, 25, 26, 35, 36, 37], "directli": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "perturb": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "speed": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "factor": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "3x150": [16, 18, 19], "hour": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "These": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "rate": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "visual": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "logdir": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "labelsmooth": 16, "someth": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "tensorflow": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "found": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "continu": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "press": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "ctrl": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36, 37], "engw8ksktzqs24zbv5dgcg": 16, "22t11": 16, "scan": [16, 18, 19, 21, 23, 31, 35, 36], "116068": 16, "scalar": [16, 18, 19, 21, 23, 31, 35, 36], "listen": [16, 18, 19, 23, 31, 35, 36], "url": [16, 18, 19, 21, 23, 25, 26, 31, 35, 36], "xxxx": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "text": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "saw": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "consol": [16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "typic": [16, 18, 19, 21], "avoid": [16, 18, 21], "commonli": [16, 18, 19, 21, 24, 28, 29, 31], "nbest": [16, 21, 26], "scale": [16, 21, 24, 26, 28, 29], "lattic": [16, 19, 21, 23, 24, 28, 29, 36, 37], "score": [16, 21, 23, 36, 37], "uniqu": [16, 21, 23, 36, 37], "pkufool": [16, 19, 24], "icefall_asr_aishell_conformer_ctc": 16, "transcrib": [16, 18, 19, 21], "lang_char": [16, 18], "token": [16, 18, 19, 21, 24, 28, 29, 31], "word": [16, 18, 19, 21, 24, 28, 29, 31], "bac009s0764w0121": [16, 18, 19], "bac009s0764w0122": [16, 18, 19], "bac009s0764w0123": [16, 18, 19], "tran": [16, 19, 21, 24, 28, 29], "graph": [16, 19, 21, 23, 24, 28, 29, 36, 37], "id": [16, 19, 21, 24, 28, 29], "conveni": [16, 19, 21], "eo": [16, 19, 21], "easili": [16, 19, 21], "obtain": [16, 18, 19, 21, 24, 28, 29], "84": 16, "list": [16, 18, 19, 21, 24, 28, 29], "soxi": [16, 18, 19, 21, 24, 31], "sampl": [16, 18, 19, 21, 24, 25, 31, 36, 37], "precis": [16, 18, 19, 21, 23, 24, 31, 36, 37], "bit": [16, 18, 19, 21, 24, 31], "04": [16, 18, 19, 21, 24, 28, 29], "67263": [16, 18, 19], "315": [16, 18, 19, 21, 24], "cdda": [16, 18, 19, 21, 24, 31], "sector": [16, 18, 19, 21, 24, 31], "135k": [16, 18, 19], "256k": [16, 18, 19, 21], "sign": [16, 18, 19, 21, 31], "integ": [16, 18, 19, 21, 31], "pcm": [16, 18, 19, 21, 31], "65840": [16, 18, 19], "625": [16, 18, 19], "132k": [16, 18, 19], "64000": [16, 18, 19], "300": [16, 18, 19, 21, 23, 36], "128k": [16, 18, 19, 31], "displai": [16, 18, 19, 21], "topologi": [16, 21], "07": [16, 18, 19, 21], "53": [16, 23, 24, 29, 35, 36], "707": [16, 21], "229": 16, "attention_dim": [16, 18, 21], "num_decoder_lay": [16, 21], "vgg_frontend": [16, 18, 21], "use_feat_batchnorm": [16, 21], "f2fd997f752ed11bbef4c306652c433e83f9cf12": 16, "sun": 16, "sep": 16, "46": [16, 21], "33cfe45": 16, "d57a873": 16, "wed": [16, 18, 21], "nov": [16, 21], "hw": 16, "kangwei": 16, "icefall_aishell3": 16, "k2_releas": 16, "tokens_fil": 16, "words_fil": [16, 21, 31], "num_path": [16, 21, 23, 36, 37], "ngram_lm_scal": [16, 21], "attention_decoder_scal": [16, 21], "nbest_scal": [16, 21], "sos_id": [16, 21], "eos_id": [16, 21], "num_class": [16, 21, 31], "4336": [16, 18], "708": [16, 18, 21, 31], "242": [16, 21], "131": [16, 21], "134": 16, "269": [16, 28, 29], "275": 16, "241": 16, "293": [16, 21], "704": [16, 28], "369": [16, 21], "\u751a": [16, 18], "\u81f3": [16, 18], "\u51fa": [16, 18], "\u73b0": [16, 18], "\u4ea4": [16, 18], "\u6613": [16, 18], "\u51e0": [16, 18], "\u4e4e": [16, 18], "\u505c": [16, 18], "\u6b62": 16, "\u7684": [16, 18, 19], "\u60c5": [16, 18], "\u51b5": [16, 18], "\u4e00": [16, 18], "\u4e8c": [16, 18], "\u7ebf": [16, 18, 19], "\u57ce": [16, 18], "\u5e02": [16, 18], "\u867d": [16, 18], "\u7136": [16, 18], "\u4e5f": [16, 18, 19], "\u5904": [16, 18], "\u4e8e": [16, 18], "\u8c03": [16, 18], "\u6574": [16, 18], "\u4e2d": [16, 18, 19], "\u4f46": [16, 18, 19], "\u56e0": [16, 18], "\u4e3a": [16, 18], "\u805a": [16, 18], "\u96c6": [16, 18], "\u4e86": [16, 18, 19], "\u8fc7": [16, 18], "\u591a": [16, 18], "\u516c": [16, 18], "\u5171": [16, 18], "\u8d44": [16, 18], "\u6e90": [16, 18], "371": 16, "37": [16, 18, 21, 28], "38": [16, 18, 21, 28], "683": 16, "47": [16, 21], "651": [16, 31], "654": 16, "659": 16, "752": 16, "321": 16, "887": 16, "340": 16, "370": 16, "\u751a\u81f3": [16, 19], "\u51fa\u73b0": [16, 19], "\u4ea4\u6613": [16, 19], "\u51e0\u4e4e": [16, 19], "\u505c\u6b62": 16, "\u60c5\u51b5": [16, 19], "\u4e00\u4e8c": [16, 19], "\u57ce\u5e02": [16, 19], "\u867d\u7136": [16, 19], "\u5904\u4e8e": [16, 19], "\u8c03\u6574": [16, 19], "\u56e0\u4e3a": [16, 19], "\u805a\u96c6": [16, 19], "\u8fc7\u591a": [16, 19], "\u516c\u5171": [16, 19], "\u8d44\u6e90": [16, 19], "372": 16, "recor": [16, 21], "highest": [16, 21], "965": 16, "966": 16, "821": 16, "822": 16, "826": 16, "916": 16, "115": [16, 21], "345": 16, "888": 16, "889": 16, "limit": [16, 18, 21, 33, 36], "memori": [16, 18, 21, 33], "upgrad": [16, 21], "pro": [16, 21], "finish": [16, 18, 19, 21, 23, 24, 28, 29, 31, 36, 37], "deploi": [16, 21], "At": [16, 21], "doe": [16, 18, 21, 31], "checkout": [16, 21], "v2": [16, 21], "cmake": [16, 21, 35], "dcmake_build_typ": [16, 21, 35], "j": [16, 21], "hlg_decod": [16, 21], "four": [16, 21], "messag": [16, 21, 23, 25, 26, 35, 36, 37], "nn_model": [16, 21], "use_gpu": [16, 21], "word_tabl": [16, 21], "caution": [16, 21], "forward": [16, 21, 25], "cu": [16, 21], "int": [16, 21], "char": [16, 21], "124": [16, 21], "98": 16, "142": [16, 19, 21], "150": [16, 21], "693": [16, 28], "165": [16, 21], "nnet_output": [16, 21], "182": [16, 24], "180": [16, 21], "489": 16, "45": [16, 18, 21], "216": [16, 21, 28, 29], "mandarin": 17, "corpu": 17, "beij": 17, "shell": 17, "technologi": 17, "ltd": 17, "400": 17, "peopl": 17, "accent": 17, "area": 17, "china": 17, "invit": 17, "particip": 17, "conduct": 17, "quiet": 17, "indoor": 17, "high": 17, "fidel": 17, "microphon": 17, "downsampl": 17, "16khz": 17, "manual": 17, "accuraci": 17, "through": 17, "profession": 17, "annot": 17, "strict": 17, "inspect": 17, "free": [17, 35], "academ": 17, "moder": 17, "amount": 17, "research": 17, "field": 17, "openslr": 17, "ctc": [17, 20, 22, 26, 27, 30], "stateless": [17, 20, 23, 35, 36, 37], "instead": [18, 36], "rnn": [18, 23, 25, 35, 36, 37], "As": [18, 21], "head": [18, 33], "dim": [18, 23, 36], "layer": [18, 23, 33, 35, 36, 37], "feedforward": [18, 23, 36], "embed": [18, 23, 35, 36, 37], "conv1d": [18, 23, 35, 36, 37], "kernel": 18, "left": [18, 36, 37], "context": [18, 23, 33, 35, 36, 37], "nn": [18, 23, 25, 26, 35, 36, 37], "tanh": 18, "linear": 18, "borrow": 18, "ieeexplor": 18, "ieee": 18, "stamp": 18, "jsp": 18, "arnumb": 18, "9054419": 18, "predict": [18, 23, 35, 36, 37], "modif": 18, "right": [18, 33, 36], "charact": 18, "unit": 18, "vocabulari": 18, "87939824": 18, "88": 18, "optimized_transduc": 18, "extra": [18, 33, 36], "technqiu": 18, "propos": [18, 33, 37], "improv": 18, "end": [18, 23, 25, 26, 31, 35, 36, 37], "furthermor": 18, "maximum": 18, "emit": 18, "per": [18, 23, 36, 37], "frame": [18, 23, 25, 36, 37], "simplifi": [18, 33], "significantli": 18, "degrad": 18, "exactli": 18, "benchmark": 18, "unprun": 18, "advantag": 18, "minim": 18, "pruned_transducer_stateless": [18, 23, 33, 36], "altern": 18, "though": 18, "transducer_stateless_modifi": 18, "option": [18, 24, 28, 29, 31], "pr": 18, "gb": 18, "ram": 18, "small": [18, 28, 29, 31], "tri": 18, "prob": [18, 35], "appli": [18, 33], "configur": [18, 24, 28, 29, 31], "c": [18, 19, 23, 25, 26, 31, 35, 36, 37], "lagz6hrcqxoigbfd5e0y3q": 18, "03t14": 18, "8477": 18, "sym": [18, 23, 36, 37], "beam_search": [18, 23, 36, 37], "decoding_method": 18, "beam_4": 18, "28": [18, 21, 24], "ensur": 18, "give": 18, "poor": 18, "531": [18, 19], "994": [18, 21], "176": [18, 21], "027": 18, "encoder_out_dim": 18, "f4fefe4882bc0ae59af951da3f47335d5495ef71": 18, "feb": 18, "50d2281": 18, "mar": 18, "0815224919": 18, "75d558775b": 18, "mmnv8": 18, "72": [18, 21], "878": [18, 29], "257": [18, 28, 29], "880": 18, "267": [18, 28, 29], "891": 18, "__floordiv__": 18, "length": [18, 36, 37], "x_len": 18, "163": [18, 21], "320": 18, "\u6ede": 18, "322": 18, "759": 18, "760": 18, "919": 18, "922": 18, "046": 18, "047": 18, "319": [18, 21], "214": [18, 21], "215": [18, 21, 24], "402": 18, "topk_hyp_index": 18, "topk_index": 18, "logit": 18, "583": [18, 29], "2000": 19, "lji9mwuorlow3jkdhxwk8a": 19, "13t11": 19, "4454": 19, "icefall_asr_aishell_tdnn_lstm_ctc": 19, "858": [19, 21], "389": [19, 21], "154": 19, "161": [19, 21], "536": 19, "171": [19, 21, 28, 29], "539": 19, "917": 19, "207": [19, 21], "129": 19, "\u505c\u6ede": 19, "222": [19, 21], "statelessx": [20, 22, 32, 33, 34], "zipform": [20, 22, 32, 34], "mmi": [20, 22], "blank": [20, 22], "skip": [20, 22, 23, 35, 36, 37], "ligru": [20, 27], "full": [21, 23, 25, 26, 35, 36, 37], "libri": [21, 23, 25, 26, 35, 36, 37], "960": [21, 23, 25, 26, 35, 36, 37], "subset": [21, 23, 25, 26, 35, 36, 37], "3x960": [21, 23, 25, 26, 35, 36, 37], "2880": [21, 23, 25, 26, 35, 36, 37], "lzgnetjwrxc3yghnmd4kpw": 21, "24t16": 21, "43": 21, "4540": 21, "sentenc": 21, "piec": 21, "And": [21, 23, 25, 26, 35, 36, 37], "neither": 21, "nor": 21, "vocab": 21, "work": 21, "5000": 21, "44": [21, 28, 29], "033": 21, "538": 21, "full_libri": 21, "406": 21, "464": 21, "548": 21, "776": 21, "652": [21, 31], "109226120": 21, "714": [21, 28], "473": 21, "944": 21, "1328": 21, "54": [21, 24, 28, 29], "443": [21, 24], "2563": 21, "56": [21, 28], "494": 21, "592": 21, "331": [21, 24], "1715": 21, "52576": 21, "1424": 21, "807": 21, "506": 21, "808": [21, 28], "522": 21, "362": 21, "565": 21, "1477": 21, "106": 21, "2922": 21, "208": 21, "4295": 21, "52343": 21, "396": 21, "3584": 21, "433": 21, "680": [21, 28], "_pickl": 21, "unpicklingerror": 21, "kei": 21, "hlg_modifi": 21, "g_4_gram": [21, 24, 28, 29], "106000": [21, 24], "496": [21, 24], "875": [21, 24], "212k": 21, "267440": [21, 24], "1253": [21, 24], "535k": 21, "83": [21, 24], "77200": [21, 24], "361": [21, 24], "154k": 21, "554": 21, "260": 21, "7178d67e594bc7fa89c2b331ad7bd1c62a6a9eb4": 21, "tue": 21, "8d93169": 21, "266": [21, 24], "268": [21, 24], "601": 21, "758": 21, "025": 21, "204": 21, "425": 21, "broffel": 21, "osom": 21, "427": 21, "723": 21, "775": 21, "881": 21, "352": 21, "234": 21, "384": 21, "whole": [21, 24, 28, 29, 36, 37], "ngram": [21, 24, 28, 29], "857": 21, "979": 21, "980": 21, "055": 21, "117": 21, "051": 21, "363": 21, "959": [21, 29], "546": 21, "598": 21, "599": [21, 24], "833": 21, "834": 21, "915": 21, "076": 21, "110": 21, "397": 21, "999": [21, 23, 36, 37], "concaten": 21, "bucket": 21, "sampler": 21, "1000": 21, "ctc_decod": 21, "ngram_lm_rescor": 21, "attention_rescor": 21, "kind": [21, 23, 25, 26, 35, 36, 37], "316": 21, "118": 21, "221": 21, "125": [21, 31], "136": 21, "144": 21, "159": [21, 31], "543": 21, "174": 21, "topo": 21, "547": 21, "729": 21, "111": 21, "702": 21, "703": 21, "545": 21, "122": 21, "280": 21, "135": [21, 31], "153": [21, 31], "945": 21, "475": 21, "191": [21, 28, 29], "398": 21, "199": [21, 24], "515": 21, "205": 21, "w": [21, 28, 29], "deseri": 21, "441": 21, "fsaclass": 21, "loadfsa": 21, "const": 21, "string": 21, "c10": 21, "ignor": 21, "dummi": 21, "589": 21, "attention_scal": 21, "656": 21, "162": 21, "169": [21, 28, 29], "188": 21, "624": 21, "519": [21, 29], "632": 21, "645": [21, 31], "243": 21, "970": 21, "303": 21, "179": 21, "suitabl": [23, 35, 36, 37], "pruned_transducer_stateless2": [23, 33, 36], "pruned_transducer_stateless4": [23, 33, 36], "pruned_transducer_stateless5": [23, 33, 36], "scroll": [23, 25, 26, 35, 36, 37], "scratch": [23, 25, 26, 35, 36, 37], "paper": [23, 35, 36, 37], "arxiv": [23, 35, 36, 37], "ab": [23, 35, 36, 37], "2206": [23, 35, 36, 37], "13236": [23, 35, 36, 37], "rework": [23, 33, 36], "daniel": [23, 36, 37], "joint": [23, 35, 36, 37], "contrari": [23, 35, 36, 37], "convent": [23, 35, 36, 37], "recurr": [23, 35, 36, 37], "fp16": [23, 25, 26, 35, 36, 37], "half": [23, 36, 37], "2x": [23, 36, 37], "dimens": [23, 36, 37], "littl": [23, 36], "allow": [23, 36], "436000": [23, 25, 26, 35, 36, 37], "438000": [23, 25, 26, 35, 36, 37], "qogspbgsr8kzcrmmie9jgw": 23, "20t15": [23, 35, 36], "4468": [23, 35, 36], "210171": [23, 35, 36], "access": [23, 25, 26, 35, 36, 37], "6008": [23, 25, 26, 35, 36, 37], "localhost": [23, 25, 26, 35, 36, 37], "expos": [23, 25, 26, 35, 36, 37], "proxi": [23, 25, 26, 35, 36, 37], "bind_al": [23, 25, 26, 35, 36, 37], "suggest": [23, 25, 26, 35, 36, 37], "both": [23, 25, 26, 35, 36, 37], "lowest": [23, 25, 26, 35, 36, 37], "fast_beam_search": [23, 25, 35, 36, 37], "474000": [23, 35, 36, 37], "largest": [23, 36, 37], "posterior": [23, 25, 36, 37], "algorithm": [23, 36, 37], "pdf": [23, 26, 36, 37], "1211": [23, 36, 37], "3711": [23, 36, 37], "espnet": [23, 36, 37], "net": [23, 36, 37], "beam_search_transduc": [23, 36, 37], "basicli": [23, 36, 37], "topk": [23, 36, 37], "expand": [23, 36, 37], "mode": [23, 36, 37], "being": [23, 36, 37], "hardcod": [23, 36, 37], "composit": [23, 36, 37], "between": [23, 36, 37], "log_prob": [23, 36, 37], "hard": [23, 33, 36, 37], "2211": [23, 36, 37], "00484": [23, 36, 37], "rnnt": [23, 36, 37], "effici": [23, 36, 37], "fast_beam_search_lg": [23, 36, 37], "trivial": [23, 36, 37], "fast_beam_search_nbest": [23, 36, 37], "random_path": [23, 36, 37], "shortest": [23, 36, 37], "fast_beam_search_nbest_lg": [23, 36, 37], "logic": [23, 36, 37], "includ": [23, 25, 26, 35, 36, 37], "But": [23, 25, 26, 35, 36, 37], "smallest": [23, 35, 36, 37], "icefall_asr_librispeech_tdnn": 24, "lstm_ctc": 24, "flac": 24, "116k": 24, "140k": 24, "343k": 24, "164k": 24, "105k": 24, "174k": 24, "pretraind": 24, "168": 24, "170": 24, "581": 24, "584": [24, 29], "209": 24, "791": 24, "245": 24, "099": 24, "methond": [24, 28, 29], "725": 24, "403": 24, "631": 24, "010": 24, "guidanc": 25, "calcul": [25, 36, 37], "bigger": 25, "threshold": 25, "simpli": 25, "discard": 25, "prevent": 25, "convolut": [25, 33, 36], "similar": [25, 36, 37], "lconv": 25, "encourag": [25, 26, 35], "stabil": [25, 26], "doesn": 25, "warm": [25, 26], "pruned_transducer_stateless7_ctc_b": 25, "xyozukpeqm62hbilud4upa": [25, 26], "ctc_guild_decode_b": 25, "pretrained_ctc": 25, "jit_pretrained_ctc": 25, "yfyeung": 25, "wechat": 26, "zipformer_mmi": 26, "worker": [26, 35], "hp": 26, "zengwei": [26, 35], "tdnn_ligru_ctc": 28, "enough": [28, 29, 31], "luomingshuang": [28, 29], "icefall_asr_timit_tdnn_ligru_ctc": 28, "pretrained_average_9_25": 28, "fdhc0_si1559": [28, 29], "felc0_si756": [28, 29], "fmgd0_si1564": [28, 29], "ffprobe": [28, 29], "show_format": [28, 29], "nistspher": [28, 29], "database_id": [28, 29], "database_vers": [28, 29], "utterance_id": [28, 29], "dhc0_si1559": [28, 29], "sample_min": [28, 29], "4176": [28, 29], "sample_max": [28, 29], "5984": [28, 29], "bitrat": [28, 29], "258": [28, 29], "audio": [28, 29], "pcm_s16le": [28, 29], "hz": [28, 29], "s16": [28, 29], "256": [28, 29], "elc0_si756": [28, 29], "1546": [28, 29], "1989": [28, 29], "mgd0_si1564": [28, 29], "7626": [28, 29], "10573": [28, 29], "660": 28, "183": [28, 29], "695": 28, "697": 28, "210": [28, 29], "829": 28, "sil": [28, 29], "dh": [28, 29], "ih": [28, 29], "uw": [28, 29], "ah": [28, 29], "ii": [28, 29], "z": [28, 29], "aa": [28, 29], "ei": [28, 29], "dx": [28, 29], "uh": [28, 29], "ng": [28, 29], "th": [28, 29], "eh": [28, 29], "jh": [28, 29], "er": [28, 29], "ai": [28, 29], "hh": [28, 29], "aw": 28, "ae": [28, 29], "705": 28, "715": 28, "720": 28, "251": [28, 29], "348": 28, "ch": 28, "icefall_asr_timit_tdnn_lstm_ctc": 29, "pretrained_average_16_25": 29, "816": 29, "827": 29, "387": 29, "unk": 29, "739": 29, "971": 29, "977": 29, "978": 29, "981": 29, "ow": 29, "ykubhb5wrmosxykid1z9eg": 31, "23t23": 31, "sinc": [31, 35], "icefall_asr_yesno_tdnn": 31, "l_disambig": 31, "lexicon_disambig": 31, "arpa": 31, "0_0_0_1_0_0_0_1": 31, "0_0_1_0_0_0_1_0": 31, "0_0_1_0_0_1_1_1": 31, "0_0_1_0_1_0_0_1": 31, "0_0_1_1_0_0_0_1": 31, "0_0_1_1_0_1_1_0": 31, "0_0_1_1_1_0_0_0": 31, "0_0_1_1_1_1_0_0": 31, "0_1_0_0_0_1_0_0": 31, "0_1_0_0_1_0_1_0": 31, "0_1_0_1_0_0_0_0": 31, "0_1_0_1_1_1_0_0": 31, "0_1_1_0_0_1_1_1": 31, "0_1_1_1_0_0_1_0": 31, "0_1_1_1_1_0_1_0": 31, "1_0_0_0_0_0_0_0": 31, "1_0_0_0_0_0_1_1": 31, "1_0_0_1_0_1_1_1": 31, "1_0_1_1_0_1_1_1": 31, "1_0_1_1_1_1_0_1": 31, "1_1_0_0_0_1_1_1": 31, "1_1_0_0_1_0_1_1": 31, "1_1_0_1_0_1_0_0": 31, "1_1_0_1_1_0_0_1": 31, "1_1_0_1_1_1_1_0": 31, "1_1_1_0_0_1_0_1": 31, "1_1_1_0_1_0_1_0": 31, "1_1_1_1_0_0_1_0": 31, "1_1_1_1_1_0_0_0": 31, "1_1_1_1_1_1_1_1": 31, "54080": 31, "507": 31, "108k": 31, "No": 31, "ye": 31, "hebrew": 31, "NO": 31, "621": 31, "119": 31, "127": 31, "650": 31, "139": 31, "143": 31, "198": 31, "181": 31, "186": 31, "187": 31, "287": 31, "correctli": 31, "simplest": 31, "former": 33, "idea": 33, "achiev": 33, "mask": [33, 36, 37], "wenet": 33, "did": 33, "argument": 33, "adapt": 33, "complic": 33, "techniqu": 33, "bank": 33, "compon": 33, "memor": 33, "histori": 33, "introduc": 33, "variant": 33, "pruned_stateless_emformer_rnnt2": 33, "conv_emformer_transducer_stateless": 33, "convemform": 33, "ourself": 33, "mechan": 33, "conv_emformer_transducer_stateless2": 33, "onlin": 35, "lstm_transducer_stateless": 35, "architectur": 35, "lower": 35, "prepare_giga_speech": 35, "cj2vtpiwqhkn9q1tx6ptpg": 35, "hidden": 35, "1024": 35, "pnnx": 35, "submodul": 35, "updat": 35, "recurs": 35, "init": 35, "dncnn_python": 35, "dncnn_build_benchmark": 35, "off": 35, "dncnn_build_exampl": 35, "dncnn_build_tool": 35, "j4": 35, "pwd": 35, "quantiz": 35, "abl": 35, "ncnn2int8": 35, "third": 35, "param": 35, "extern": 35, "stateless2": 35, "compar": 36, "dynam": [36, 37], "chunk": [36, 37], "causal": 36, "short": [36, 37], "2012": 36, "05481": 36, "flag": 36, "indic": [36, 37], "whether": 36, "must": 36, "sequenc": [36, 37], "uniformli": [36, 37], "seen": [36, 37], "97vkxf80ru61cnp2alwzzg": 36, "streaming_decod": [36, 37], "acoust": [36, 37], "wise": [36, 37], "subsampl": [36, 37], "equal": [36, 37], "where": 36, "parallel": [36, 37], "bath": [36, 37], "parallelli": [36, 37], "seem": 36, "benefit": 36, "might": [36, 37], "mismatch": 36, "yourself": [36, 37], "mdoel": 36, "pruned_transducer_stateless7_stream": 37, "len": 37, "320m": 37, "550": 37, "scriptmodul": 37, "jit_trace_export": 37, "jit_trace_pretrain": 37, "task": 38}, "objects": {}, "objtypes": {}, "objnames": {}, "titleterms": {"follow": 0, "code": 0, "style": 0, "contribut": [1, 3], "document": 1, "how": [2, 10, 12, 13, 14], "creat": [2, 9], "recip": [2, 38], "data": [2, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "prepar": [2, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "train": [2, 6, 9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "decod": [2, 9, 10, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "pre": [2, 6, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "model": [2, 6, 10, 12, 13, 14, 15, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "frequent": 4, "ask": 4, "question": 4, "faq": 4, "oserror": 4, "libtorch_hip": 4, "so": 4, "cannot": 4, "open": 4, "share": 4, "object": 4, "file": 4, "directori": 4, "attributeerror": 4, "modul": 4, "distutil": 4, "ha": 4, "attribut": 4, "version": 4, "huggingfac": [5, 7], "space": 7, "youtub": [7, 9], "video": [7, 9], "icefal": [8, 9], "content": [8, 38], "instal": [9, 16, 18, 19, 21, 24, 28, 29], "0": 9, "pytorch": 9, "torchaudio": 9, "1": [9, 16, 18, 19, 21], "k2": 9, "2": [9, 16, 18, 19, 21], "lhots": 9, "3": [9, 16, 18, 21], "download": [9, 16, 18, 19, 21, 23, 24, 25, 26, 28, 29, 31, 35, 36, 37], "exampl": [9, 16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "virtual": 9, "environ": 9, "activ": 9, "your": 9, "4": 9, "5": 9, "test": 9, "export": [10, 11, 12, 13, 14, 15, 23, 25, 26, 35, 36, 37], "state_dict": [10, 23, 25, 26, 35, 36, 37], "when": [10, 12, 13, 14], "us": [10, 12, 13, 14, 23, 25, 26, 35, 36, 37], "run": 10, "py": 10, "ncnn": [11, 35], "onnx": 12, "torch": [13, 14, 23, 25, 26, 35, 36, 37], "jit": [13, 14, 23, 25, 26, 35, 36, 37], "script": [13, 23, 25, 26, 36, 37], "trace": [14, 35, 37], "conform": [16, 21, 33], "ctc": [16, 19, 21, 24, 25, 28, 29, 31], "configur": [16, 19, 21, 23, 25, 26, 35, 36, 37], "option": [16, 19, 21, 23, 25, 26, 35, 36, 37], "log": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "usag": [16, 18, 19, 21, 23, 25, 26, 35, 36, 37], "case": [16, 18, 19, 21], "kaldifeat": [16, 18, 19, 21, 24, 28, 29, 31], "hlg": [16, 19, 21], "attent": [16, 21], "rescor": [16, 21], "colab": [16, 18, 19, 21, 24, 28, 29, 31], "notebook": [16, 18, 19, 21, 24, 28, 29, 31], "deploy": [16, 21], "c": [16, 21], "aishel": 17, "stateless": 18, "transduc": [18, 23, 35, 36, 37], "The": 18, "loss": 18, "todo": 18, "greedi": 18, "search": 18, "beam": 18, "modifi": 18, "tdnn": [19, 24, 28, 29, 31], "lstm": [19, 24, 29, 35], "non": 20, "stream": [20, 32, 33, 36, 37], "asr": [20, 32], "lm": 21, "comput": 21, "wer": 21, "n": 21, "gram": 21, "librispeech": [22, 34], "prune": [23, 36], "statelessx": [23, 36], "pretrain": [23, 25, 26, 35, 36, 37], "deploi": [23, 36, 37], "sherpa": [23, 36, 37], "infer": [24, 28, 29, 31], "zipform": [25, 26, 37], "blank": 25, "skip": 25, "mmi": 26, "timit": 27, "ligru": 28, "yesno": 30, "introduct": 33, "emform": 33, "which": 35, "simul": [36, 37], "real": [36, 37], "tabl": 38}, "envversion": {"sphinx.domains.c": 2, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 8, "sphinx.domains.index": 1, "sphinx.domains.javascript": 2, "sphinx.domains.math": 2, "sphinx.domains.python": 3, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.todo": 2, "sphinx": 57}, "alltitles": {"Follow the code style": [[0, "follow-the-code-style"]], "Contributing to Documentation": [[1, "contributing-to-documentation"]], "How to create a recipe": [[2, "how-to-create-a-recipe"]], "Data Preparation": [[2, "data-preparation"], [18, "data-preparation"]], "Training": [[2, "training"], [9, "training"], [16, "training"], [18, "training"], [19, "training"], [21, "training"], [23, "training"], [24, "training"], [25, "training"], [26, "training"], [28, "training"], [29, "training"], [31, "training"], [35, "training"], [36, "training"], [37, "training"]], "Decoding": [[2, "decoding"], [9, "decoding"], [16, "decoding"], [18, "decoding"], [19, "decoding"], [21, "decoding"], [23, "decoding"], [24, "decoding"], [25, "decoding"], [26, "decoding"], [28, "decoding"], [29, "decoding"], [31, "decoding"], [35, "decoding"], [36, "decoding"], [37, "decoding"]], "Pre-trained model": [[2, "pre-trained-model"]], "Contributing": [[3, "contributing"]], "Frequently Asked Questions (FAQs)": [[4, "frequently-asked-questions-faqs"]], "OSError: libtorch_hip.so: cannot open shared object file: no such file or directory": [[4, "oserror-libtorch-hip-so-cannot-open-shared-object-file-no-such-file-or-directory"]], "AttributeError: module \u2018distutils\u2019 has no attribute \u2018version\u2019": [[4, "attributeerror-module-distutils-has-no-attribute-version"]], "Huggingface": [[5, "huggingface"]], "Pre-trained models": [[6, "pre-trained-models"]], "Huggingface spaces": [[7, "huggingface-spaces"]], "YouTube Video": [[7, "youtube-video"], [9, "youtube-video"]], "Icefall": [[8, "icefall"]], "Contents:": [[8, null]], "Installation": [[9, "installation"]], "(0) Install PyTorch and torchaudio": [[9, "install-pytorch-and-torchaudio"]], "(1) Install k2": [[9, "install-k2"]], "(2) Install lhotse": [[9, "install-lhotse"]], "(3) Download icefall": [[9, "download-icefall"]], "Installation example": [[9, "installation-example"]], "(1) Create a virtual environment": [[9, "create-a-virtual-environment"]], "(2) Activate your virtual environment": [[9, "activate-your-virtual-environment"]], "(3) Install k2": [[9, "id1"]], "(4) Install lhotse": [[9, "id2"]], "(5) Download icefall": [[9, "id3"]], "Test Your Installation": [[9, "test-your-installation"]], "Data preparation": [[9, "data-preparation"], [16, "data-preparation"], [19, "data-preparation"], [21, "data-preparation"], [23, "data-preparation"], [24, "data-preparation"], [25, "data-preparation"], [26, "data-preparation"], [28, "data-preparation"], [29, "data-preparation"], [31, "data-preparation"], [35, "data-preparation"], [36, "data-preparation"], [37, "data-preparation"]], "Export model.state_dict()": [[10, "export-model-state-dict"], [23, "export-model-state-dict"], [25, "export-model-state-dict"], [26, "export-model-state-dict"], [35, "export-model-state-dict"], [36, "export-model-state-dict"], [37, "export-model-state-dict"]], "When to use it": [[10, "when-to-use-it"], [12, "when-to-use-it"], [13, "when-to-use-it"], [14, "when-to-use-it"]], "How to export": [[10, "how-to-export"], [12, "how-to-export"], [13, "how-to-export"], [14, "how-to-export"]], "How to use the exported model": [[10, "how-to-use-the-exported-model"], [12, "how-to-use-the-exported-model"], [13, "how-to-use-the-exported-model"]], "Use the exported model to run decode.py": [[10, "use-the-exported-model-to-run-decode-py"]], "Export to ncnn": [[11, "export-to-ncnn"]], "Export to ONNX": [[12, "export-to-onnx"]], "Export model with torch.jit.script()": [[13, "export-model-with-torch-jit-script"]], "Export model with torch.jit.trace()": [[14, "export-model-with-torch-jit-trace"]], "How to use the exported models": [[14, "how-to-use-the-exported-models"]], "Model export": [[15, "model-export"]], "Conformer CTC": [[16, "conformer-ctc"], [21, "conformer-ctc"]], "Configurable options": [[16, "configurable-options"], [19, "configurable-options"], [21, "configurable-options"], [23, "configurable-options"], [25, "configurable-options"], [26, "configurable-options"], [35, "configurable-options"], [36, "configurable-options"], [37, "configurable-options"]], "Pre-configured options": [[16, "pre-configured-options"], [19, "pre-configured-options"], [21, "pre-configured-options"], [23, "pre-configured-options"], [25, "pre-configured-options"], [26, "pre-configured-options"], [35, "pre-configured-options"], [36, "pre-configured-options"], [37, "pre-configured-options"]], "Training logs": [[16, "training-logs"], [18, "training-logs"], [19, "training-logs"], [21, "training-logs"], [23, "training-logs"], [25, "training-logs"], [26, "training-logs"], [35, "training-logs"], [36, "training-logs"], [37, "training-logs"]], "Usage examples": [[16, "usage-examples"], [18, "usage-examples"], [19, "usage-examples"], [21, "usage-examples"]], "Case 1": [[16, "case-1"], [18, "case-1"], [19, "case-1"], [21, "case-1"]], "Case 2": [[16, "case-2"], [18, "case-2"], [19, "case-2"], [21, "case-2"]], "Case 3": [[16, "case-3"], [18, "case-3"], [21, "case-3"]], "Pre-trained Model": [[16, "pre-trained-model"], [18, "pre-trained-model"], [19, "pre-trained-model"], [21, "pre-trained-model"], [24, "pre-trained-model"], [28, "pre-trained-model"], [29, "pre-trained-model"], [31, "pre-trained-model"]], "Install kaldifeat": [[16, "install-kaldifeat"], [18, "install-kaldifeat"], [19, "install-kaldifeat"], [21, "install-kaldifeat"], [24, "install-kaldifeat"], [28, "install-kaldifeat"], [29, "install-kaldifeat"]], "Download the pre-trained model": [[16, "download-the-pre-trained-model"], [18, "download-the-pre-trained-model"], [19, "download-the-pre-trained-model"], [21, "download-the-pre-trained-model"], [24, "download-the-pre-trained-model"], [28, "download-the-pre-trained-model"], [29, "download-the-pre-trained-model"], [31, "download-the-pre-trained-model"]], "Usage": [[16, "usage"], [18, "usage"], [19, "usage"], [21, "usage"]], "CTC decoding": [[16, "ctc-decoding"], [21, "ctc-decoding"], [21, "id2"]], "HLG decoding": [[16, "hlg-decoding"], [16, "id2"], [19, "hlg-decoding"], [21, "hlg-decoding"], [21, "id3"]], "HLG decoding + attention decoder rescoring": [[16, "hlg-decoding-attention-decoder-rescoring"]], "Colab notebook": [[16, "colab-notebook"], [18, "colab-notebook"], [19, "colab-notebook"], [21, "colab-notebook"], [24, "colab-notebook"], [28, "colab-notebook"], [29, "colab-notebook"], [31, "colab-notebook"]], "Deployment with C++": [[16, "deployment-with-c"], [21, "deployment-with-c"]], "aishell": [[17, "aishell"]], "Stateless Transducer": [[18, "stateless-transducer"]], "The Model": [[18, "the-model"]], "The Loss": [[18, "the-loss"]], "Todo": [[18, "id1"]], "Greedy search": [[18, "greedy-search"]], "Beam search": [[18, "beam-search"]], "Modified Beam search": [[18, "modified-beam-search"]], "TDNN-LSTM CTC": [[19, "tdnn-lstm-ctc"]], "Non Streaming ASR": [[20, "non-streaming-asr"]], "HLG decoding + LM rescoring": [[21, "hlg-decoding-lm-rescoring"]], "HLG decoding + LM rescoring + attention decoder rescoring": [[21, "hlg-decoding-lm-rescoring-attention-decoder-rescoring"]], "Compute WER with the pre-trained model": [[21, "compute-wer-with-the-pre-trained-model"]], "HLG decoding + n-gram LM rescoring": [[21, "hlg-decoding-n-gram-lm-rescoring"]], "HLG decoding + n-gram LM rescoring + attention decoder rescoring": [[21, "hlg-decoding-n-gram-lm-rescoring-attention-decoder-rescoring"]], "LibriSpeech": [[22, "librispeech"], [34, "librispeech"]], "Pruned transducer statelessX": [[23, "pruned-transducer-statelessx"], [36, "pruned-transducer-statelessx"]], "Usage example": [[23, "usage-example"], [25, "usage-example"], [26, "usage-example"], [35, "usage-example"], [36, "usage-example"], [37, "usage-example"]], "Export Model": [[23, "export-model"], [36, "export-model"], [37, "export-model"]], "Export model using torch.jit.script()": [[23, "export-model-using-torch-jit-script"], [25, "export-model-using-torch-jit-script"], [26, "export-model-using-torch-jit-script"], [36, "export-model-using-torch-jit-script"], [37, "export-model-using-torch-jit-script"]], "Download pretrained models": [[23, "download-pretrained-models"], [25, "download-pretrained-models"], [26, "download-pretrained-models"], [35, "download-pretrained-models"], [36, "download-pretrained-models"], [37, "download-pretrained-models"]], "Deploy with Sherpa": [[23, "deploy-with-sherpa"], [36, "deploy-with-sherpa"], [37, "deploy-with-sherpa"]], "TDNN-LSTM-CTC": [[24, "tdnn-lstm-ctc"], [29, "tdnn-lstm-ctc"]], "Inference with a pre-trained model": [[24, "inference-with-a-pre-trained-model"], [28, "inference-with-a-pre-trained-model"], [29, "inference-with-a-pre-trained-model"], [31, "inference-with-a-pre-trained-model"]], "Zipformer CTC Blank Skip": [[25, "zipformer-ctc-blank-skip"]], "Export models": [[25, "export-models"], [26, "export-models"], [35, "export-models"]], "Zipformer MMI": [[26, "zipformer-mmi"]], "TIMIT": [[27, "timit"]], "TDNN-LiGRU-CTC": [[28, "tdnn-ligru-ctc"]], "YesNo": [[30, "yesno"]], "TDNN-CTC": [[31, "tdnn-ctc"]], "Download kaldifeat": [[31, "download-kaldifeat"]], "Streaming ASR": [[32, "streaming-asr"]], "Introduction": [[33, "introduction"]], "Streaming Conformer": [[33, "streaming-conformer"]], "Streaming Emformer": [[33, "streaming-emformer"]], "LSTM Transducer": [[35, "lstm-transducer"]], "Which model to use": [[35, "which-model-to-use"]], "Export model using torch.jit.trace()": [[35, "export-model-using-torch-jit-trace"], [37, "export-model-using-torch-jit-trace"]], "Export model for ncnn": [[35, "export-model-for-ncnn"]], "Simulate streaming decoding": [[36, "simulate-streaming-decoding"], [37, "simulate-streaming-decoding"]], "Real streaming decoding": [[36, "real-streaming-decoding"], [37, "real-streaming-decoding"]], "Zipformer Transducer": [[37, "zipformer-transducer"]], "Recipes": [[38, "recipes"]], "Table of Contents": [[38, null]]}, "indexentries": {}})