Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: openai/whisper-medium
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: stt
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# stt
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3390
|
| 21 |
+
- Wer Ortho: 24.0940
|
| 22 |
+
- Wer: 16.4814
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 16
|
| 43 |
+
- eval_batch_size: 16
|
| 44 |
+
- seed: 42
|
| 45 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
+
- lr_scheduler_type: cosine
|
| 47 |
+
- lr_scheduler_warmup_steps: 50
|
| 48 |
+
- training_steps: 8000
|
| 49 |
+
- mixed_precision_training: Native AMP
|
| 50 |
+
-
|
| 51 |
+
### Training results
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|
| 53 |
+
|:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|
|
| 54 |
+
| 0.4135 | 0.6180 | 500 | 0.4069 | 29.9115 | 21.6319 |
|
| 55 |
+
| 0.2036 | 1.2361 | 1000 | 0.3584 | 25.8738 | 18.3552 |
|
| 56 |
+
| 0.1899 | 1.8541 | 1500 | 0.3390 | 24.0940 | 16.4814 |
|
| 57 |
+
| 0.0978 | 2.4722 | 2000 | 0.3406 | 24.1957 | 16.8982 |
|
| 58 |
+
| 0.0584 | 3.0902 | 2500 | 0.3589 | 22.7718 | 15.9189 |
|
| 59 |
+
| 0.0457 | 3.7083 | 3000 | 0.3660 | 23.3075 | 15.8580 |
|
| 60 |
+
| 0.0203 | 4.3263 | 3500 | 0.3762 | 22.9108 | 15.7394 |
|
| 61 |
+
| 0.0193 | 4.9444 | 4000 | 0.3683 | 22.0192 | 15.2616 |
|
| 62 |
+
| 0.0073 | 5.5624 | 4500 | 0.3926 | 22.5447 | 15.5801 |
|
| 63 |
+
| 0.0022 | 6.1805 | 5000 | 0.4065 | 21.5649 | 14.9092 |
|
| 64 |
+
| 0.0022 | 6.7985 | 5500 | 0.4080 | 21.2835 | 14.6313 |
|
| 65 |
+
| 0.0009 | 7.4166 | 6000 | 0.4180 | 21.2564 | 14.6415 |
|
| 66 |
+
| 0.0007 | 8.0346 | 6500 | 0.4244 | 21.2361 | 14.6551 |
|
| 67 |
+
| 0.0006 | 8.6527 | 7000 | 0.4283 | 21.3276 | 14.6957 |
|
| 68 |
+
| 0.0006 | 9.2707 | 7500 | 0.4297 | 21.3378 | 14.7059 |
|
| 69 |
+
| 0.0006 | 9.8888 | 8000 | 0.4300 | 21.3276 | 14.7093 |
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
- Transformers 4.51.3
|
| 73 |
+
- Pytorch 2.5.1+cu121
|
| 74 |
+
- Datasets 3.6.0
|
| 75 |
+
- Tokenizers 0.21.1
|