Edit model card

Pre-trained Conformer-CTC models for the librispeech dataset with icefall.

The model was trained on full LibriSpeech with the scripts in icefall.
See (https://github.com/k2-fsa/icefall/pull/13) for more details of this model.

How to use

See (https://github.com/k2-fsa/icefall/blob/master/egs/librispeech/ASR/conformer_ctc/README.md)

Training procedure

The version of the mainly repositories are list below.
k2: https://github.com/k2-fsa/k2/commit/81cec9ec736d2c603ad75d933bb3e3a3706fb0dd
icefall: https://github.com/k2-fsa/icefall/commit/ef233486ae6d21bacb940de45efb35d0c334605c
lhotse: https://github.com/lhotse-speech/lhotse/commit/5dfe0f4c02b1334ebb7db6d67e1141fe406ca76b

git clone https://github.com/k2-fsa/icefall
cd icefall
git checkout ef233486
  • Preparing data.
cd egs/librispeech/ASR
bash ./prepare.sh
  • Training
export CUDA_VISIBLE_DEVICES="0,1,2,3"
python conformer_ctc/train.py --bucketing-sampler True \                                
                                --concatenate-cuts False \
                                --max-duration 200 \
                                --full-libri True \
                                --world-size 4

Evaluation results

The best decoding results (WERs) on LibriSpeech test-clean and test-other are listed below, we got this results by averaging models from epoch 15 to 34.

test-clean test-other
WER 2.57% 5.94%
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference API
Unable to determine this model's library. Check the docs .

Spaces using pkufool/icefall_asr_librispeech_conformer_ctc 6