update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- wer
|
7 |
+
model-index:
|
8 |
+
- name: whisper-tiny-fluers_V2_telugu_Augmentation_full_datset_V2_e5
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# whisper-tiny-fluers_V2_telugu_Augmentation_full_datset_V2_e5
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.3374
|
20 |
+
- Wer: 61.2000
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 1.5e-05
|
40 |
+
- train_batch_size: 4
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- gradient_accumulation_steps: 8
|
44 |
+
- total_train_batch_size: 32
|
45 |
+
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
|
46 |
+
- lr_scheduler_type: linear
|
47 |
+
- lr_scheduler_warmup_steps: 500
|
48 |
+
- training_steps: 4000
|
49 |
+
- mixed_precision_training: Native AMP
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
55 |
+
| 1.2926 | 0.09 | 300 | 1.3993 | 129.5 |
|
56 |
+
| 1.0948 | 0.18 | 600 | 1.2674 | 109.1500 |
|
57 |
+
| 0.6591 | 0.28 | 900 | 0.5519 | 81.55 |
|
58 |
+
| 0.5326 | 0.37 | 1200 | 0.4361 | 72.55 |
|
59 |
+
| 0.4737 | 0.46 | 1500 | 0.4036 | 69.2000 |
|
60 |
+
| 0.4239 | 0.55 | 1800 | 0.3793 | 63.6 |
|
61 |
+
| 0.4011 | 0.64 | 2100 | 0.3625 | 62.2500 |
|
62 |
+
| 0.3687 | 0.73 | 2400 | 0.3651 | 62.5 |
|
63 |
+
| 0.3712 | 0.83 | 2700 | 0.3491 | 59.9 |
|
64 |
+
| 0.3686 | 0.92 | 3000 | 0.3438 | 60.6500 |
|
65 |
+
| 0.3381 | 1.01 | 3300 | 0.3391 | 58.25 |
|
66 |
+
| 0.3483 | 1.1 | 3600 | 0.3385 | 60.0500 |
|
67 |
+
| 0.341 | 1.19 | 3900 | 0.3374 | 61.2000 |
|
68 |
+
|
69 |
+
|
70 |
+
### Framework versions
|
71 |
+
|
72 |
+
- Transformers 4.28.0.dev0
|
73 |
+
- Pytorch 1.12.1
|
74 |
+
- Datasets 2.11.0
|
75 |
+
- Tokenizers 0.13.3
|