mirror of
https://github.com/k2-fsa/icefall.git
synced 2025-08-09 18:12:19 +00:00
d167aad4ab
* add streaming zipformer codes * add test_model.py * add export.py, pretrained.py, jit_pretrained.py * add cached_len for pooling module * add jit_trace_export.py and jit_trace_pretrained.py * fix bug in jit.trace * update RESULTS.md * add CI test * minor fix in pruned_transducer_stateless7/zipformer.py * update README.md
55 lines
4.8 KiB
Markdown
55 lines
4.8 KiB
Markdown
# Introduction
|
|
|
|
Please refer to <https://icefall.readthedocs.io/en/latest/recipes/librispeech/index.html> for how to run models in this recipe.
|
|
|
|
[./RESULTS.md](./RESULTS.md) contains the latest results.
|
|
|
|
# Transducers
|
|
|
|
There are various folders containing the name `transducer` in this folder.
|
|
The following table lists the differences among them.
|
|
|
|
| | Encoder | Decoder | Comment |
|
|
|---------------------------------------|---------------------|--------------------|---------------------------------------------------|
|
|
| `transducer` | Conformer | LSTM | |
|
|
| `transducer_stateless` | Conformer | Embedding + Conv1d | Using optimized_transducer from computing RNN-T loss |
|
|
| `transducer_stateless2` | Conformer | Embedding + Conv1d | Using torchaudio for computing RNN-T loss |
|
|
| `transducer_lstm` | LSTM | LSTM | |
|
|
| `transducer_stateless_multi_datasets` | Conformer | Embedding + Conv1d | Using data from GigaSpeech as extra training data |
|
|
| `pruned_transducer_stateless` | Conformer | Embedding + Conv1d | Using k2 pruned RNN-T loss |
|
|
| `pruned_transducer_stateless2` | Conformer(modified) | Embedding + Conv1d | Using k2 pruned RNN-T loss |
|
|
| `pruned_transducer_stateless3` | Conformer(modified) | Embedding + Conv1d | Using k2 pruned RNN-T loss + using GigaSpeech as extra training data |
|
|
| `pruned_transducer_stateless4` | Conformer(modified) | Embedding + Conv1d | same as pruned_transducer_stateless2 + save averaged models periodically during training + delay penalty |
|
|
| `pruned_transducer_stateless5` | Conformer(modified) | Embedding + Conv1d | same as pruned_transducer_stateless4 + more layers + random combiner|
|
|
| `pruned_transducer_stateless6` | Conformer(modified) | Embedding + Conv1d | same as pruned_transducer_stateless4 + distillation with hubert|
|
|
| `pruned_transducer_stateless7` | Zipformer | Embedding + Conv1d | First experiment with Zipformer from Dan|
|
|
| `pruned_transducer_stateless7_ctc` | Zipformer | Embedding + Conv1d | Same as pruned_transducer_stateless7, but with extra CTC head|
|
|
| `pruned_transducer_stateless7_ctc_bs` | Zipformer | Embedding + Conv1d | pruned_transducer_stateless7_ctc + blank skip |
|
|
| `pruned_transducer_stateless7_streaming` | Streaming Zipformer | Embedding + Conv1d | streaming version of pruned_transducer_stateless7 |
|
|
| `pruned_transducer_stateless8` | Zipformer | Embedding + Conv1d | Same as pruned_transducer_stateless7, but using extra data from GigaSpeech|
|
|
| `pruned_stateless_emformer_rnnt2` | Emformer(from torchaudio) | Embedding + Conv1d | Using Emformer from torchaudio for streaming ASR|
|
|
| `conv_emformer_transducer_stateless` | ConvEmformer | Embedding + Conv1d | Using ConvEmformer for streaming ASR + mechanisms in reworked model |
|
|
| `conv_emformer_transducer_stateless2` | ConvEmformer | Embedding + Conv1d | Using ConvEmformer with simplified memory for streaming ASR + mechanisms in reworked model |
|
|
| `lstm_transducer_stateless` | LSTM | Embedding + Conv1d | Using LSTM with mechanisms in reworked model |
|
|
| `lstm_transducer_stateless2` | LSTM | Embedding + Conv1d | Using LSTM with mechanisms in reworked model + gigaspeech (multi-dataset setup) |
|
|
| `lstm_transducer_stateless3` | LSTM | Embedding + Conv1d | Using LSTM with mechanisms in reworked model + gradient filter + delay penalty |
|
|
|
|
The decoder in `transducer_stateless` is modified from the paper
|
|
[Rnn-Transducer with Stateless Prediction Network](https://ieeexplore.ieee.org/document/9054419/).
|
|
We place an additional Conv1d layer right after the input embedding layer.
|
|
|
|
# CTC
|
|
|
|
| | Encoder | Comment |
|
|
|------------------------------|--------------------|------------------------------|
|
|
| `conformer-ctc` | Conformer | Use auxiliary attention head |
|
|
| `conformer-ctc2` | Reworked Conformer | Use auxiliary attention head |
|
|
| `conformer-ctc3` | Reworked Conformer | Streaming version + delay penalty |
|
|
|
|
# MMI
|
|
|
|
| | Encoder | Comment |
|
|
|------------------------------|-----------|---------------------------------------------------|
|
|
| `conformer-mmi` | Conformer | |
|
|
| `zipformer-mmi` | Zipformer | CTC warmup + use HP as decoding graph for decoding |
|