diff --git a/egs/aishell/ASR/conformer_ctc/decode.py b/egs/aishell/ASR/conformer_ctc/decode.py index a3e7f98e3..751b7d5b5 100755 --- a/egs/aishell/ASR/conformer_ctc/decode.py +++ b/egs/aishell/ASR/conformer_ctc/decode.py @@ -335,7 +335,7 @@ def decode_dataset( lexicon: Lexicon, sos_id: int, eos_id: int, -) -> Dict[str, List[Tuple[List[int], List[int]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -410,7 +410,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): if params.method == "attention-decoder": # Set it to False since there are too many logs. diff --git a/egs/aishell/ASR/conformer_mmi/decode.py b/egs/aishell/ASR/conformer_mmi/decode.py index ac68b61e7..4db367e36 100755 --- a/egs/aishell/ASR/conformer_mmi/decode.py +++ b/egs/aishell/ASR/conformer_mmi/decode.py @@ -347,7 +347,7 @@ def decode_dataset( lexicon: Lexicon, sos_id: int, eos_id: int, -) -> Dict[str, List[Tuple[List[int], List[int]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -422,7 +422,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): if params.method == "attention-decoder": # Set it to False since there are too many logs. diff --git a/egs/aishell/ASR/pruned_transducer_stateless2/decode.py b/egs/aishell/ASR/pruned_transducer_stateless2/decode.py index 3a68e8765..a12934d55 100755 --- a/egs/aishell/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/aishell/ASR/pruned_transducer_stateless2/decode.py @@ -326,7 +326,7 @@ def decode_dataset( model: nn.Module, token_table: k2.SymbolTable, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -396,7 +396,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell/ASR/pruned_transducer_stateless3/decode.py b/egs/aishell/ASR/pruned_transducer_stateless3/decode.py index 3268c8bb2..d159e420b 100755 --- a/egs/aishell/ASR/pruned_transducer_stateless3/decode.py +++ b/egs/aishell/ASR/pruned_transducer_stateless3/decode.py @@ -340,7 +340,7 @@ def decode_dataset( model: nn.Module, token_table: k2.SymbolTable, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -410,7 +410,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell/ASR/tdnn_lstm_ctc/decode.py b/egs/aishell/ASR/tdnn_lstm_ctc/decode.py index 45c1c4ec1..66b734fc4 100755 --- a/egs/aishell/ASR/tdnn_lstm_ctc/decode.py +++ b/egs/aishell/ASR/tdnn_lstm_ctc/decode.py @@ -208,7 +208,7 @@ def decode_dataset( model: nn.Module, HLG: k2.Fsa, lexicon: Lexicon, -) -> Dict[str, List[Tuple[List[int], List[int]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -274,7 +274,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell/ASR/transducer_stateless/decode.py b/egs/aishell/ASR/transducer_stateless/decode.py index d78821b95..780b0c4bb 100755 --- a/egs/aishell/ASR/transducer_stateless/decode.py +++ b/egs/aishell/ASR/transducer_stateless/decode.py @@ -264,7 +264,7 @@ def decode_dataset( params: AttributeDict, model: nn.Module, lexicon: Lexicon, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -328,7 +328,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell/ASR/transducer_stateless_modified-2/decode.py b/egs/aishell/ASR/transducer_stateless_modified-2/decode.py index a9dca995f..ea3f94fd8 100755 --- a/egs/aishell/ASR/transducer_stateless_modified-2/decode.py +++ b/egs/aishell/ASR/transducer_stateless_modified-2/decode.py @@ -304,7 +304,7 @@ def decode_dataset( model: nn.Module, token_table: k2.SymbolTable, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -374,7 +374,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell/ASR/transducer_stateless_modified/decode.py b/egs/aishell/ASR/transducer_stateless_modified/decode.py index 9e827e1d1..65fcda873 100755 --- a/egs/aishell/ASR/transducer_stateless_modified/decode.py +++ b/egs/aishell/ASR/transducer_stateless_modified/decode.py @@ -308,7 +308,7 @@ def decode_dataset( model: nn.Module, token_table: k2.SymbolTable, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -378,7 +378,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell2/ASR/pruned_transducer_stateless5/decode.py b/egs/aishell2/ASR/pruned_transducer_stateless5/decode.py index 7d6f6f6d5..915737f4a 100755 --- a/egs/aishell2/ASR/pruned_transducer_stateless5/decode.py +++ b/egs/aishell2/ASR/pruned_transducer_stateless5/decode.py @@ -478,7 +478,7 @@ def decode_dataset( lexicon: Lexicon, graph_compiler: CharCtcTrainingGraphCompiler, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -547,7 +547,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/aishell4/ASR/pruned_transducer_stateless5/decode.py b/egs/aishell4/ASR/pruned_transducer_stateless5/decode.py index 739dfc4a1..14e44c7d9 100755 --- a/egs/aishell4/ASR/pruned_transducer_stateless5/decode.py +++ b/egs/aishell4/ASR/pruned_transducer_stateless5/decode.py @@ -342,7 +342,7 @@ def decode_dataset( model: nn.Module, lexicon: Lexicon, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -410,7 +410,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/alimeeting/ASR/pruned_transducer_stateless2/decode.py b/egs/alimeeting/ASR/pruned_transducer_stateless2/decode.py index 65fc74728..6358fe970 100755 --- a/egs/alimeeting/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/alimeeting/ASR/pruned_transducer_stateless2/decode.py @@ -331,7 +331,7 @@ def decode_dataset( model: nn.Module, lexicon: Lexicon, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -399,7 +399,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/gigaspeech/ASR/conformer_ctc/decode.py b/egs/gigaspeech/ASR/conformer_ctc/decode.py index f4b438aad..51406667e 100755 --- a/egs/gigaspeech/ASR/conformer_ctc/decode.py +++ b/egs/gigaspeech/ASR/conformer_ctc/decode.py @@ -173,13 +173,13 @@ def get_params() -> AttributeDict: def post_processing( - results: List[Tuple[List[str], List[str]]], -) -> List[Tuple[List[str], List[str]]]: + results: List[Tuple[str, List[str], List[str]]], +) -> List[Tuple[str, List[str], List[str]]]: new_results = [] - for ref, hyp in results: + for key, ref, hyp in results: new_ref = asr_text_post_processing(" ".join(ref)).split() new_hyp = asr_text_post_processing(" ".join(hyp)).split() - new_results.append((new_ref, new_hyp)) + new_results.append((key, new_ref, new_hyp)) return new_results @@ -408,7 +408,7 @@ def decode_dataset( sos_id: int, eos_id: int, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -502,7 +502,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[str], List[str]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): if params.method == "attention-decoder": # Set it to False since there are too many logs. diff --git a/egs/gigaspeech/ASR/pruned_transducer_stateless2/decode.py b/egs/gigaspeech/ASR/pruned_transducer_stateless2/decode.py index d7ecc3fdc..5849a3471 100755 --- a/egs/gigaspeech/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/gigaspeech/ASR/pruned_transducer_stateless2/decode.py @@ -203,13 +203,13 @@ def get_parser(): def post_processing( - results: List[Tuple[List[str], List[str]]], -) -> List[Tuple[List[str], List[str]]]: + results: List[Tuple[str, List[str], List[str]]], +) -> List[Tuple[str, List[str], List[str]]]: new_results = [] - for ref, hyp in results: + for key, ref, hyp in results: new_ref = asr_text_post_processing(" ".join(ref)).split() new_hyp = asr_text_post_processing(" ".join(hyp)).split() - new_results.append((new_ref, new_hyp)) + new_results.append((key, new_ref, new_hyp)) return new_results @@ -340,7 +340,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -407,7 +407,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/conformer_ctc/decode.py b/egs/librispeech/ASR/conformer_ctc/decode.py index 7a8fb2130..3f3b1acda 100755 --- a/egs/librispeech/ASR/conformer_ctc/decode.py +++ b/egs/librispeech/ASR/conformer_ctc/decode.py @@ -480,7 +480,7 @@ def decode_dataset( sos_id: int, eos_id: int, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -577,7 +577,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[int], List[int]]]], ): if params.method in ("attention-decoder", "rnn-lm"): # Set it to False since there are too many logs. diff --git a/egs/librispeech/ASR/conformer_ctc2/decode.py b/egs/librispeech/ASR/conformer_ctc2/decode.py index 6b9da12a9..97f2f2d39 100755 --- a/egs/librispeech/ASR/conformer_ctc2/decode.py +++ b/egs/librispeech/ASR/conformer_ctc2/decode.py @@ -587,7 +587,7 @@ def decode_dataset( sos_id: int, eos_id: int, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -684,7 +684,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): if params.method in ("attention-decoder", "rnn-lm"): # Set it to False since there are too many logs. diff --git a/egs/librispeech/ASR/conformer_mmi/decode.py b/egs/librispeech/ASR/conformer_mmi/decode.py index 23372034a..fc9861489 100755 --- a/egs/librispeech/ASR/conformer_mmi/decode.py +++ b/egs/librispeech/ASR/conformer_mmi/decode.py @@ -404,7 +404,7 @@ def decode_dataset( sos_id: int, eos_id: int, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -487,7 +487,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): if params.method == "attention-decoder": # Set it to False since there are too many logs. diff --git a/egs/librispeech/ASR/conv_emformer_transducer_stateless/decode.py b/egs/librispeech/ASR/conv_emformer_transducer_stateless/decode.py index a03fe2684..620d69a19 100755 --- a/egs/librispeech/ASR/conv_emformer_transducer_stateless/decode.py +++ b/egs/librispeech/ASR/conv_emformer_transducer_stateless/decode.py @@ -366,7 +366,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -436,7 +436,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/conv_emformer_transducer_stateless2/decode.py b/egs/librispeech/ASR/conv_emformer_transducer_stateless2/decode.py index d204a9d75..98b8290b5 100755 --- a/egs/librispeech/ASR/conv_emformer_transducer_stateless2/decode.py +++ b/egs/librispeech/ASR/conv_emformer_transducer_stateless2/decode.py @@ -366,7 +366,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -436,7 +436,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/lstm_transducer_stateless/decode.py b/egs/librispeech/ASR/lstm_transducer_stateless/decode.py index bfc158e0a..27414d717 100755 --- a/egs/librispeech/ASR/lstm_transducer_stateless/decode.py +++ b/egs/librispeech/ASR/lstm_transducer_stateless/decode.py @@ -496,7 +496,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -570,7 +570,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/lstm_transducer_stateless2/decode.py b/egs/librispeech/ASR/lstm_transducer_stateless2/decode.py index 21ae563cb..420202cad 100755 --- a/egs/librispeech/ASR/lstm_transducer_stateless2/decode.py +++ b/egs/librispeech/ASR/lstm_transducer_stateless2/decode.py @@ -497,7 +497,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -571,7 +571,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/lstm_transducer_stateless3/decode.py b/egs/librispeech/ASR/lstm_transducer_stateless3/decode.py index 9842bf00d..5be23c50c 100755 --- a/egs/librispeech/ASR/lstm_transducer_stateless3/decode.py +++ b/egs/librispeech/ASR/lstm_transducer_stateless3/decode.py @@ -496,7 +496,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -570,7 +570,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_stateless_emformer_rnnt2/decode.py b/egs/librispeech/ASR/pruned_stateless_emformer_rnnt2/decode.py index 282ce3737..2d5724d30 100755 --- a/egs/librispeech/ASR/pruned_stateless_emformer_rnnt2/decode.py +++ b/egs/librispeech/ASR/pruned_stateless_emformer_rnnt2/decode.py @@ -354,7 +354,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -424,7 +424,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless/decode.py index ab6cf336c..b11fb960a 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless/decode.py @@ -511,7 +511,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[str, Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -585,7 +585,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[str, Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless2/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless2/decode.py index 34fd31e7e..7852dafc9 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless2/decode.py @@ -534,7 +534,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -608,7 +608,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless3/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless3/decode.py index 72d6f656c..fa4f1e7d9 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless3/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless3/decode.py @@ -781,7 +781,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless4/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless4/decode.py index 8431492e6..b75a72a15 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless4/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless4/decode.py @@ -538,7 +538,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -612,7 +612,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[str, Tuple[List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless5/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless5/decode.py index 32bbd16f7..96103500b 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless5/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless5/decode.py @@ -524,7 +524,7 @@ def decode_dataset( sp: spm.SentencePieceProcessor, word_table: Optional[k2.SymbolTable] = None, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -598,7 +598,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/pruned_transducer_stateless6/decode.py b/egs/librispeech/ASR/pruned_transducer_stateless6/decode.py index 2f69ba401..74df04006 100755 --- a/egs/librispeech/ASR/pruned_transducer_stateless6/decode.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless6/decode.py @@ -350,7 +350,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -420,7 +420,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/tdnn_lstm_ctc/decode.py b/egs/librispeech/ASR/tdnn_lstm_ctc/decode.py index f1aacb5e7..7d0cd0bf3 100755 --- a/egs/librispeech/ASR/tdnn_lstm_ctc/decode.py +++ b/egs/librispeech/ASR/tdnn_lstm_ctc/decode.py @@ -274,7 +274,7 @@ def decode_dataset( HLG: k2.Fsa, lexicon: Lexicon, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -345,7 +345,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/transducer/decode.py b/egs/librispeech/ASR/transducer/decode.py index 83e924256..24f243974 100755 --- a/egs/librispeech/ASR/transducer/decode.py +++ b/egs/librispeech/ASR/transducer/decode.py @@ -261,7 +261,7 @@ def decode_dataset( params: AttributeDict, model: nn.Module, sp: spm.SentencePieceProcessor, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -327,7 +327,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[str, Tuple[List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/transducer_lstm/decode.py b/egs/librispeech/ASR/transducer_lstm/decode.py index 43debe643..604235e2a 100755 --- a/egs/librispeech/ASR/transducer_lstm/decode.py +++ b/egs/librispeech/ASR/transducer_lstm/decode.py @@ -258,7 +258,7 @@ def decode_dataset( params: AttributeDict, model: nn.Module, sp: spm.SentencePieceProcessor, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -324,7 +324,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/transducer_stateless/decode.py b/egs/librispeech/ASR/transducer_stateless/decode.py index 19a685090..74bba9cad 100755 --- a/egs/librispeech/ASR/transducer_stateless/decode.py +++ b/egs/librispeech/ASR/transducer_stateless/decode.py @@ -313,7 +313,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -383,7 +383,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/transducer_stateless2/decode.py b/egs/librispeech/ASR/transducer_stateless2/decode.py index f48ce82f4..ac2807241 100755 --- a/egs/librispeech/ASR/transducer_stateless2/decode.py +++ b/egs/librispeech/ASR/transducer_stateless2/decode.py @@ -313,7 +313,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -383,7 +383,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/librispeech/ASR/transducer_stateless_multi_datasets/decode.py b/egs/librispeech/ASR/transducer_stateless_multi_datasets/decode.py index 2bb6df5d6..d596e05cb 100755 --- a/egs/librispeech/ASR/transducer_stateless_multi_datasets/decode.py +++ b/egs/librispeech/ASR/transducer_stateless_multi_datasets/decode.py @@ -314,7 +314,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -384,7 +384,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/spgispeech/ASR/pruned_transducer_stateless2/decode.py b/egs/spgispeech/ASR/pruned_transducer_stateless2/decode.py index 722cd8dbd..c39bd0530 100755 --- a/egs/spgispeech/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/spgispeech/ASR/pruned_transducer_stateless2/decode.py @@ -328,7 +328,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -398,7 +398,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() test_set_cers = dict() diff --git a/egs/tal_csasr/ASR/pruned_transducer_stateless5/decode.py b/egs/tal_csasr/ASR/pruned_transducer_stateless5/decode.py index 45f702163..b624913f5 100755 --- a/egs/tal_csasr/ASR/pruned_transducer_stateless5/decode.py +++ b/egs/tal_csasr/ASR/pruned_transducer_stateless5/decode.py @@ -413,7 +413,7 @@ def decode_dataset( lexicon: Lexicon, decoding_graph: Optional[k2.Fsa] = None, sp: spm.SentencePieceProcessor = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -515,7 +515,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/tedlium3/ASR/pruned_transducer_stateless/decode.py b/egs/tedlium3/ASR/pruned_transducer_stateless/decode.py index bb352baa7..2b294e601 100755 --- a/egs/tedlium3/ASR/pruned_transducer_stateless/decode.py +++ b/egs/tedlium3/ASR/pruned_transducer_stateless/decode.py @@ -313,7 +313,7 @@ def decode_dataset( model: nn.Module, sp: spm.SentencePieceProcessor, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -383,7 +383,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/tedlium3/ASR/transducer_stateless/decode.py b/egs/tedlium3/ASR/transducer_stateless/decode.py index c1aa2c366..d3e9e55e7 100755 --- a/egs/tedlium3/ASR/transducer_stateless/decode.py +++ b/egs/tedlium3/ASR/transducer_stateless/decode.py @@ -291,7 +291,7 @@ def decode_dataset( params: AttributeDict, model: nn.Module, sp: spm.SentencePieceProcessor, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -357,7 +357,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/timit/ASR/tdnn_ligru_ctc/decode.py b/egs/timit/ASR/tdnn_ligru_ctc/decode.py index 84d9f7f1b..4f2aa2340 100644 --- a/egs/timit/ASR/tdnn_ligru_ctc/decode.py +++ b/egs/timit/ASR/tdnn_ligru_ctc/decode.py @@ -274,7 +274,7 @@ def decode_dataset( HLG: k2.Fsa, lexicon: Lexicon, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -345,7 +345,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/timit/ASR/tdnn_lstm_ctc/decode.py b/egs/timit/ASR/tdnn_lstm_ctc/decode.py index 7672a2e1d..5e7300cf2 100644 --- a/egs/timit/ASR/tdnn_lstm_ctc/decode.py +++ b/egs/timit/ASR/tdnn_lstm_ctc/decode.py @@ -273,7 +273,7 @@ def decode_dataset( HLG: k2.Fsa, lexicon: Lexicon, G: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -344,7 +344,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/wenetspeech/ASR/pruned_transducer_stateless2/decode.py b/egs/wenetspeech/ASR/pruned_transducer_stateless2/decode.py index bbd8680b2..f0c9bebec 100755 --- a/egs/wenetspeech/ASR/pruned_transducer_stateless2/decode.py +++ b/egs/wenetspeech/ASR/pruned_transducer_stateless2/decode.py @@ -455,7 +455,7 @@ def decode_dataset( lexicon: Lexicon, graph_compiler: CharCtcTrainingGraphCompiler, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -524,7 +524,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/wenetspeech/ASR/pruned_transducer_stateless5/decode.py b/egs/wenetspeech/ASR/pruned_transducer_stateless5/decode.py index c36df2458..344e31283 100755 --- a/egs/wenetspeech/ASR/pruned_transducer_stateless5/decode.py +++ b/egs/wenetspeech/ASR/pruned_transducer_stateless5/decode.py @@ -425,7 +425,7 @@ def decode_dataset( model: nn.Module, lexicon: Lexicon, decoding_graph: Optional[k2.Fsa] = None, -) -> Dict[str, List[Tuple[List[str], List[str]]]]: +) -> Dict[str, List[Tuple[str, List[str], List[str]]]]: """Decode dataset. Args: @@ -493,7 +493,7 @@ def decode_dataset( def save_results( params: AttributeDict, test_set_name: str, - results_dict: Dict[str, List[Tuple[List[int], List[int]]]], + results_dict: Dict[str, List[Tuple[str, List[str], List[str]]]], ): test_set_wers = dict() for key, results in results_dict.items(): diff --git a/egs/yesno/ASR/tdnn/decode.py b/egs/yesno/ASR/tdnn/decode.py index 79adcb14e..9d4ab4b61 100755 --- a/egs/yesno/ASR/tdnn/decode.py +++ b/egs/yesno/ASR/tdnn/decode.py @@ -147,7 +147,7 @@ def decode_dataset( model: nn.Module, HLG: k2.Fsa, word_table: k2.SymbolTable, -) -> List[Tuple[List[int], List[int]]]: +) -> List[Tuple[str, List[str], List[str]]]: """Decode dataset. Args: @@ -210,7 +210,7 @@ def decode_dataset( def save_results( exp_dir: Path, test_set_name: str, - results: List[Tuple[List[int], List[int]]], + results: List[Tuple[str, List[str], List[str]]], ) -> None: """Save results to `exp_dir`. Args: