mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-12-09 14:05:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

add HF

Browse Source
This commit is contained in:
AmirHussein96 2025-09-15 23:36:29 -04:00
parent e0593f3152
commit 592d981715
2 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
egs/iwslt22_ta/ASR/RESULTS.md
View File

@ -1,6 +1,7 @@
# Results
### IWSLT Tunisian training results (Stateless Pruned Transducer)
#### 2023-06-01
@ -8,7 +9,7 @@
| Decoding method | dev WER | test WER | comment |
|------------------------------------|------------|------------|------------------------------------------|
| modified beam search | 47.6 | 51.2 | --epoch 20, --avg 10 |
| modified beam search | 47.6 | 51.2 | --epoch 20, --avg 13 |
The training command for reproducing is given below:
@ -55,13 +56,13 @@ done
#### 2023-06-01
You can find a pretrained model, training logs, decoding logs, and decoding results at:
https://tensorboard.dev/experiment/yLE399ZPTzePG8B39jRyOw/
<https://huggingface.co/AmirHussein/zipformer-iwslt22-Ta>
| Decoding method | dev WER | test WER | comment |
|------------------------------------|------------|------------|------------------------------------------|
| modified beam search | 40.8 | 44.4 | --epoch 20, --avg 10 |
| modified beam search | 40.8 | 44.1 | --epoch 20, --avg 13 |
To reproduce the above result, use the following commands for training:
@ -70,7 +71,7 @@ To reproduce the above result, use the following commands for training:
```
export CUDA_VISIBLE_DEVICES="0,1"
./zipformer/train.py \
--world-size 2 \
--world-size 4 \
--num-epochs 20 \
--start-epoch 1 \
--use-fp16 1 \
@ -99,7 +100,7 @@ for method in modified_beam_search; do
--num-encoder-layers 2,2,2,2,2,2 \
--feedforward-dim 512,768,1024,1536,1024,768 \
--encoder-dim 192,256,384,512,384,256 \
--encoder-unmasked-dim 192,192,256,256,256,192
--encoder-unmasked-dim 192,192,256,256,256,192 \
--use-averaged-model true
done
```

3
egs/iwslt22_ta/ST/RESULTS.md
View File

@ -58,6 +58,7 @@ done
#### 2023-06-01
You can find a pretrained model, training logs, decoding logs, and decoding results at:
<https://huggingface.co/AmirHussein/zipformer-iwslt22-Ta>
@ -95,8 +96,6 @@ To reproduce the above result, use the following commands for training:
```
The tensorboard training log can be found at
https://tensorboard.dev/experiment/4sa4M1mRQyKjOE4o95mWUw/
The decoding command is: