Hugging Face's logo

Sonal0205
/
whisper-medium-te

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Telugu
whisper
hf-asr-leaderboard
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
Sonal0205 commited on
Commit
e8be558
•
1 Parent(s): d4836f3

End of training

Browse files
Files changed (1) hide show
  1. README.md +12 -12
README.md CHANGED
@@ -7,34 +7,34 @@ tags:
7
  - hf-asr-leaderboard
8
  - generated_from_trainer
9
  datasets:
10
- - Sonal0205/telugu_asr
11
  metrics:
12
  - wer
13
  model-index:
14
- - name: Whisper Medium te - telugu
15
  results:
16
  - task:
17
  name: Automatic Speech Recognition
18
  type: automatic-speech-recognition
19
  dataset:
20
  name: Telugu ASR
21
- type: Sonal0205/telugu_asr
22
  args: 'config: te, split: test'
23
  metrics:
24
  - name: Wer
25
  type: wer
26
- value: 39.00118906064209
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
30
  should probably proofread and complete it, then remove this comment. -->
31
 
32
- # Whisper Medium te - telugu
33
 
34
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Telugu ASR dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.1044
37
- - Wer: 39.0012
38
 
39
  ## Model description
40
 
@@ -66,15 +66,15 @@ The following hyperparameters were used during training:
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Wer |
68
  |:-------------:|:------:|:----:|:---------------:|:-------:|
69
- | 0.0415 | 2.1142 | 1000 | 0.0918 | 57.5208 |
70
- | 0.0104 | 4.2283 | 2000 | 0.0984 | 38.3769 |
71
- | 0.0023 | 6.3425 | 3000 | 0.1018 | 31.8074 |
72
- | 0.0001 | 8.4567 | 4000 | 0.1044 | 39.0012 |
73
 
74
 
75
  ### Framework versions
76
 
77
  - Transformers 4.42.0.dev0
78
- - Pytorch 2.4.0.dev20240527+cu118
79
  - Datasets 2.19.1
80
  - Tokenizers 0.19.1
 
7
  - hf-asr-leaderboard
8
  - generated_from_trainer
9
  datasets:
10
+ - Sonal0205/telugu_asr_recent
11
  metrics:
12
  - wer
13
  model-index:
14
+ - name: Whisper Medium te - telugu new
15
  results:
16
  - task:
17
  name: Automatic Speech Recognition
18
  type: automatic-speech-recognition
19
  dataset:
20
  name: Telugu ASR
21
+ type: Sonal0205/telugu_asr_recent
22
  args: 'config: te, split: test'
23
  metrics:
24
  - name: Wer
25
  type: wer
26
+ value: 23.513674197384066
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
30
  should probably proofread and complete it, then remove this comment. -->
31
 
32
+ # Whisper Medium te - telugu new
33
 
34
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Telugu ASR dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.0925
37
+ - Wer: 23.5137
38
 
39
  ## Model description
40
 
 
66
 
67
  | Training Loss | Epoch | Step | Validation Loss | Wer |
68
  |:-------------:|:------:|:----:|:---------------:|:-------:|
69
+ | 0.0452 | 1.0582 | 1000 | 0.0992 | 42.1522 |
70
+ | 0.0148 | 2.1164 | 2000 | 0.0884 | 27.2889 |
71
+ | 0.0065 | 3.1746 | 3000 | 0.0924 | 24.7919 |
72
+ | 0.0013 | 4.2328 | 4000 | 0.0925 | 23.5137 |
73
 
74
 
75
  ### Framework versions
76
 
77
  - Transformers 4.42.0.dev0
78
+ - Pytorch 2.4.0.dev20240606+cu118
79
  - Datasets 2.19.1
80
  - Tokenizers 0.19.1