Hugging Face's logo

CAYTU
/
whosper-large-v2

Automatic Speech Recognition
PEFT
Safetensors
Wolof
French
English
Generated from Trainer
multilingual
ASR
Open-Source
Eval Results
Model card Files Community
sudoping01 commited on
Commit
2db6acf
·
verified ·
1 Parent(s): c42b18b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -40,7 +40,7 @@ tags:
40
  Whosper-large-v3 is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) optimized for Wolof and French speech recognition, with improved WER and CER metrics compared to its predecessor.
41
 
42
  ## Performance Metrics
43
- - **Loss**: 0.4490 (final)
44
  - **WER (Word Error Rate)**: 0.2409
45
  - **CER (Character Error Rate)**: 0.1128
46
 
@@ -95,7 +95,7 @@ mixed_precision_training: Native AMP
95
  - Tokenizers: 0.21.0
96
 
97
  ## License
98
- Apache-2.0
99
 
100
  ## Citation
101
  ```bibtex
 
40
  Whosper-large-v3 is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) optimized for Wolof and French speech recognition, with improved WER and CER metrics compared to its predecessor.
41
 
42
  ## Performance Metrics
43
+ - **Loss**: 0.4490
44
  - **WER (Word Error Rate)**: 0.2409
45
  - **CER (Character Error Rate)**: 0.1128
46
 
 
95
  - Tokenizers: 0.21.0
96
 
97
  ## License
98
+ MIT
99
 
100
  ## Citation
101
  ```bibtex