buaaalban / icefall Goto Github PK

icefall contains ASR recipes for various datasets using https://github.com/k2-fsa/k2.

You can use https://github.com/k2-fsa/sherpa to deploy models trained with icefall.

You can try pre-trained models from within your browser without the need to download or install anything by visiting https://huggingface.co/spaces/k2-fsa/automatic-speech-recognition See https://k2-fsa.github.io/icefall/huggingface/spaces.html for more details.

Please refer to https://icefall.readthedocs.io/en/latest/installation/index.html for installation.

Please refer to https://icefall.readthedocs.io/en/latest/recipes/index.html for more information.

We provide the following recipes:

• yesno
• LibriSpeech
• Aishell
• TIMIT
• TED-LIUM3
• GigaSpeech
• Aidatatang_200zh
• WenetSpeech
• Alimeeting
• Aishell4
• TAL_CSASR

This is the simplest ASR recipe in icefall and can be run on CPU. Training takes less than 30 seconds and gives you the following WER:

[test_set] %WER 0.42% [1 / 240, 0 ins, 1 del, 0 sub ]

We do provide a Colab notebook for this recipe.

Please see https://github.com/k2-fsa/icefall/blob/master/egs/librispeech/ASR/RESULTS.md for the latest results.

We provide 4 models for this recipe:

• conformer CTC model
• TDNN LSTM CTC model
• Transducer: Conformer encoder + LSTM decoder
• Transducer: Conformer encoder + Embedding decoder

The best WER we currently have is:

We provide a Colab notebook to run a pre-trained conformer CTC model:

The WER for this model is:

We provide a Colab notebook to run a pre-trained TDNN LSTM CTC model:

Using Conformer as encoder and LSTM as decoder.

The best WER with greedy search is:

We provide a Colab notebook to run a pre-trained RNN-T conformer model:

Using Conformer as encoder. The decoder consists of 1 embedding layer and 1 convolutional layer.

The best WER using modified beam search with beam size 4 is:

Note: No auxiliary losses are used in the training and no LMs are used in the decoding.

We provide a Colab notebook to run a pre-trained transducer conformer + stateless decoder model:

We provide two models for this recipe: conformer CTC model and TDNN LSTM CTC model.

The best CER we currently have is:

We provide a Colab notebook to run a pre-trained conformer CTC model: ![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg

The best CER we currently have is:

We provide a Colab notebook to run a pre-trained TransducerStateless model:

The CER for this model is:

We provide a Colab notebook to run a pre-trained TDNN LSTM CTC model:

We provide two models for this recipe: TDNN LSTM CTC model and TDNN LiGRU CTC model.

The best PER we currently have is:

We provide a Colab notebook to run a pre-trained TDNN LSTM CTC model:

The PER for this model is:

We provide a Colab notebook to run a pre-trained TDNN LiGRU CTC model:

We provide two models for this recipe: Transducer Stateless: Conformer encoder + Embedding decoder and Pruned Transducer Stateless: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

The best WER using modified beam search with beam size 4 is:

Note: No auxiliary losses are used in the training and no LMs are used in the decoding.

We provide a Colab notebook to run a pre-trained Transducer Stateless model:

The best WER using modified beam search with beam size 4 is:

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model:

We provide two models for this recipe: Conformer CTC model and Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

We provide one model for this recipe: Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model:

We provide some models for this recipe: Pruned stateless RNN-T_2: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss and Pruned stateless RNN-T_5: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

Streaming:

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless2 model:

We provide one model for this recipe: Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model:

We provide one model for this recipe: Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

The best CER(%) results:

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model:

We provide one model for this recipe: Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss.

The best results for Chinese CER(%) and English WER(%) respectivly (zh: Chinese, en: English):

We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model:

Once you have trained a model in icefall, you may want to deploy it with C++, without Python dependencies.

Please refer to the documentation https://icefall.readthedocs.io/en/latest/recipes/librispeech/conformer_ctc.html#deployment-with-c for how to do this.

We also provide a Colab notebook, showing you how to run a torch scripted model in k2 with C++. Please see: