Hugging Face's logo

steja
/
whisper-small-shona

Automatic Speech Recognition
Transformers
PyTorch
Safetensors
whisper
whisper-event
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
2
steja commited on
Commit
4c17a7b
•
1 Parent(s): 6e00f59

Model save

Browse files
Files changed (2) hide show
  1. README.md +19 -18
  2. model.safetensors +1 -1
README.md CHANGED
@@ -2,39 +2,38 @@
2
  license: apache-2.0
3
  base_model: openai/whisper-small
4
  tags:
5
- - whisper-event
6
  - generated_from_trainer
7
  datasets:
8
- - google/fleurs
9
  metrics:
10
  - wer
11
  model-index:
12
- - name: Whisper small shona
13
  results:
14
  - task:
15
  name: Automatic Speech Recognition
16
  type: automatic-speech-recognition
17
  dataset:
18
- name: google/fleurs sn_zw
19
- type: google/fleurs
20
  config: sn_zw
21
  split: test
22
  args: sn_zw
23
  metrics:
24
  - name: Wer
25
  type: wer
26
- value: 45.9
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 small shona
33
 
34
- This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the google/fleurs sn_zw dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.9653
37
- - Wer: 45.9
38
 
39
  ## Model description
40
 
@@ -58,8 +57,10 @@ The following hyperparameters were used during training:
58
  - eval_batch_size: 16
59
  - seed: 42
60
  - distributed_type: multi-GPU
 
61
  - gradient_accumulation_steps: 2
62
- - total_train_batch_size: 16
 
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
  - lr_scheduler_warmup_steps: 500
@@ -68,13 +69,13 @@ The following hyperparameters were used during training:
68
 
69
  ### Training results
70
 
71
- | Training Loss | Epoch | Step | Validation Loss | Wer |
72
- |:-------------:|:-----:|:----:|:---------------:|:-------:|
73
- | 0.09 | 8.16 | 400 | 0.8572 | 50.6875 |
74
- | 0.0019 | 16.33 | 800 | 0.9069 | 46.4438 |
75
- | 0.0008 | 24.49 | 1200 | 0.9473 | 46.0438 |
76
- | 0.0006 | 32.65 | 1600 | 0.9653 | 45.9 |
77
- | 0.0005 | 40.82 | 2000 | 0.9720 | 45.9438 |
78
 
79
 
80
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: openai/whisper-small
4
  tags:
 
5
  - generated_from_trainer
6
  datasets:
7
+ - fleurs
8
  metrics:
9
  - wer
10
  model-index:
11
+ - name: whisper-small-shona
12
  results:
13
  - task:
14
  name: Automatic Speech Recognition
15
  type: automatic-speech-recognition
16
  dataset:
17
+ name: fleurs
18
+ type: fleurs
19
  config: sn_zw
20
  split: test
21
  args: sn_zw
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 50.04219409282701
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
  should probably proofread and complete it, then remove this comment. -->
30
 
31
+ # whisper-small-shona
32
 
33
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the fleurs dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 1.1298
36
+ - Wer: 50.0422
37
 
38
  ## Model description
39
 
 
57
  - eval_batch_size: 16
58
  - seed: 42
59
  - distributed_type: multi-GPU
60
+ - num_devices: 3
61
  - gradient_accumulation_steps: 2
62
+ - total_train_batch_size: 48
63
+ - total_eval_batch_size: 48
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
  - lr_scheduler_warmup_steps: 500
 
69
 
70
  ### Training results
71
 
72
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
73
+ |:-------------:|:------:|:----:|:---------------:|:-------:|
74
+ | 0.0064 | 24.24 | 400 | 0.9630 | 50.7233 |
75
+ | 0.001 | 48.48 | 800 | 1.0617 | 49.9397 |
76
+ | 0.0005 | 72.73 | 1200 | 1.1016 | 49.9397 |
77
+ | 0.0004 | 96.97 | 1600 | 1.1220 | 49.9096 |
78
+ | 0.0003 | 121.21 | 2000 | 1.1298 | 50.0422 |
79
 
80
 
81
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d88fd19badbe8082a2e10e0914e009ad94033e6db8ec7d1e204c3d18aafb15fa
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:504c60ac53eb63b5e753c78138215f059d114086ce4f21e80b52fbf82aadc3bd
3
  size 966995080