update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- wer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: whisper-medium-NST-uf-linlr
|
| 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-medium-NST-uf-linlr
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3007
|
| 20 |
+
- Wer: 9.1220
|
| 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: 1e-05
|
| 40 |
+
- train_batch_size: 72
|
| 41 |
+
- eval_batch_size: 16
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 1000
|
| 46 |
+
- training_steps: 20000
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|
|
| 53 |
+
| 0.2046 | 0.05 | 1000 | 0.3426 | 15.2794 |
|
| 54 |
+
| 0.148 | 0.1 | 2000 | 0.3284 | 10.8324 |
|
| 55 |
+
| 0.121 | 0.15 | 3000 | 0.3092 | 12.8848 |
|
| 56 |
+
| 0.1089 | 0.2 | 4000 | 0.2808 | 10.4903 |
|
| 57 |
+
| 0.0976 | 0.25 | 5000 | 0.2617 | 9.9202 |
|
| 58 |
+
| 0.0901 | 0.3 | 6000 | 0.2604 | 21.8928 |
|
| 59 |
+
| 0.0834 | 0.35 | 7000 | 0.2877 | 9.3501 |
|
| 60 |
+
| 0.0825 | 0.4 | 8000 | 0.2794 | 9.3501 |
|
| 61 |
+
| 0.0553 | 1.05 | 9000 | 0.2845 | 9.5781 |
|
| 62 |
+
| 0.0472 | 1.1 | 10000 | 0.2814 | 24.1733 |
|
| 63 |
+
| 0.0409 | 1.15 | 11000 | 0.3084 | 8.0958 |
|
| 64 |
+
| 0.041 | 1.2 | 12000 | 0.2865 | 9.2360 |
|
| 65 |
+
| 0.0353 | 1.25 | 13000 | 0.2828 | 6.4994 |
|
| 66 |
+
| 0.0348 | 1.3 | 14000 | 0.2708 | 7.5257 |
|
| 67 |
+
| 0.0349 | 1.35 | 15000 | 0.2842 | 23.0331 |
|
| 68 |
+
| 0.0361 | 1.4 | 16000 | 0.2769 | 10.1482 |
|
| 69 |
+
| 0.0249 | 2.04 | 17000 | 0.2935 | 8.8940 |
|
| 70 |
+
| 0.0204 | 2.09 | 18000 | 0.2874 | 12.4287 |
|
| 71 |
+
| 0.0175 | 2.14 | 19000 | 0.2882 | 12.9989 |
|
| 72 |
+
| 0.0197 | 2.19 | 20000 | 0.3007 | 9.1220 |
|
| 73 |
+
|
| 74 |
+
|
| 75 |
+
### Framework versions
|
| 76 |
+
|
| 77 |
+
- Transformers 4.25.0.dev0
|
| 78 |
+
- Pytorch 1.13.0+cu117
|
| 79 |
+
- Datasets 2.6.1
|
| 80 |
+
- Tokenizers 0.13.1
|