Hugging Face's logo

geninhu
/
xls-asr-vi-40h-1B

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Vietnamese
wav2vec2
common-voice
hf-asr-leaderboard
robust-speech-event
Eval Results
Inference Endpoints
Model card Files Metrics Community
1
geninhu commited on
Commit
aec508f
•
1 Parent(s): fb03fee

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -33,20 +33,20 @@ model-index:
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on 40 hours of FPT Open Speech Dataset (FOSD).
34
 
35
  ### Benchmark WER result:
36
- | | [VIVOS](https://huggingface.co/datasets/vivos) | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) |
37
  |---|---|---|
38
  |without LM| 25.93 | 34.21 |
39
- |with 4-grams LM| 24.11 | 28.27 |
40
 
41
  ### Benchmark CER result:
42
  | | [VIVOS](https://huggingface.co/datasets/vivos) | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) |
43
  |---|---|---|
44
  |without LM| 9.24 | 19.94 |
45
- |with 4-grams LM| 10.37 | 14.42 |
46
 
47
  ## Evaluation
48
 
49
- Please use eval.py file to run evaluation
50
 
51
  ```python
52
  python eval_custom.py --model_id geninhu/xls-asr-vi-40h-1B --dataset mozilla-foundation/common_voice_7_0 --config vi --split test --log_outputs
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on 40 hours of FPT Open Speech Dataset (FOSD).
34
 
35
  ### Benchmark WER result:
36
+ | | [VIVOS](https://huggingface.co/datasets/vivos) | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) | [COMMON VOICE 8.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0)
37
  |---|---|---|
38
  |without LM| 25.93 | 34.21 |
39
+ |with 4-grams LM| 24.11 | 28.27 | 31.158 |
40
 
41
  ### Benchmark CER result:
42
  | | [VIVOS](https://huggingface.co/datasets/vivos) | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) |
43
  |---|---|---|
44
  |without LM| 9.24 | 19.94 |
45
+ |with 4-grams LM| 10.37 | 14.42 | 16.179 |
46
 
47
  ## Evaluation
48
 
49
+ Please use the eval.py file to run the evaluation
50
 
51
  ```python
52
  python eval_custom.py --model_id geninhu/xls-asr-vi-40h-1B --dataset mozilla-foundation/common_voice_7_0 --config vi --split test --log_outputs