Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/models-cards#model-card-metadata)

<<<<<<< HEAD Note: This recipe is trained with the codes from this PR https://github.com/k2-fsa/icefall/pull/261 And the SpecAugment codes from this PR https://github.com/lhotse-speech/lhotse/pull/604.

Pre-trained Transducer-Stateless models for the TEDLium3 dataset with icefall.

The model was trained on full TEDLium3 with the scripts in icefall.

Training procedure

The main repositories are list below, we will update the training and decoding scripts with the update of version.
k2: https://github.com/k2-fsa/k2 icefall: https://github.com/k2-fsa/icefall lhotse: https://github.com/lhotse-speech/lhotse

git clone https://github.com/k2-fsa/icefall
cd icefall
  • Preparing data.
cd egs/tedlium3/ASR
bash ./prepare.sh
  • Training
export CUDA_VISIBLE_DEVICES="0,1,2,3"
./pruned_transducer_stateless/train.py \
                  --world-size 4 \
                  --num-epochs 30 \
                  --start-epoch 0 \
                  --exp-dir pruned_transducer_stateless/exp \
                  --max-duration 300

Evaluation results

The decoding results (WER%) on TEDLium3 (dev and test) are listed below, we got this result by averaging models from epoch 17 to 29. The WERs are

dev test comment
greedy search 7.27 6.69 --epoch 29, --avg 13, --max-duration 100
beam search (beam size 4) 6.70 6.04 --epoch 29, --avg 13, --max-duration 100
modified beam search (beam size 4) 6.77 6.14 --epoch 29, --avg 13, --max-duration 100
Downloads last month
0
Hosted inference API

Unable to determine this model’s pipeline type. Check the docs .