Commit
·
21627c1
1
Parent(s):
ca81c07
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: openai/whisper-small
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- dataset_whisper
|
| 8 |
+
metrics:
|
| 9 |
+
- wer
|
| 10 |
+
model-index:
|
| 11 |
+
- name: Transcriber-small
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Automatic Speech Recognition
|
| 15 |
+
type: automatic-speech-recognition
|
| 16 |
+
dataset:
|
| 17 |
+
name: dataset_whisper
|
| 18 |
+
type: dataset_whisper
|
| 19 |
+
config: default
|
| 20 |
+
split: test
|
| 21 |
+
args: default
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Wer
|
| 24 |
+
type: wer
|
| 25 |
+
value: 97.23577235772358
|
| 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 |
+
# Transcriber-small
|
| 32 |
+
|
| 33 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the dataset_whisper dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 3.0153
|
| 36 |
+
- Wer: 97.2358
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 1e-05
|
| 56 |
+
- train_batch_size: 2
|
| 57 |
+
- eval_batch_size: 2
|
| 58 |
+
- seed: 42
|
| 59 |
+
- gradient_accumulation_steps: 8
|
| 60 |
+
- total_train_batch_size: 16
|
| 61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
+
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_steps: 500
|
| 64 |
+
- training_steps: 4000
|
| 65 |
+
|
| 66 |
+
### Training results
|
| 67 |
+
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 69 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 70 |
+
| 2.6006 | 4.02 | 100 | 2.6681 | 99.9350 |
|
| 71 |
+
| 1.6004 | 8.04 | 200 | 2.1138 | 107.2846 |
|
| 72 |
+
| 1.0072 | 12.06 | 300 | 1.9609 | 129.9187 |
|
| 73 |
+
| 0.5229 | 16.08 | 400 | 2.0901 | 119.0894 |
|
| 74 |
+
| 0.2155 | 20.1 | 500 | 2.2948 | 105.9187 |
|
| 75 |
+
| 0.0743 | 24.12 | 600 | 2.3731 | 100.6829 |
|
| 76 |
+
| 0.0292 | 28.14 | 700 | 2.5375 | 118.0813 |
|
| 77 |
+
| 0.0169 | 32.16 | 800 | 2.5601 | 108.0650 |
|
| 78 |
+
| 0.0121 | 36.18 | 900 | 2.6491 | 102.7642 |
|
| 79 |
+
| 0.008 | 40.2 | 1000 | 2.6436 | 94.3415 |
|
| 80 |
+
| 0.0046 | 44.22 | 1100 | 2.7131 | 89.8211 |
|
| 81 |
+
| 0.0021 | 48.24 | 1200 | 2.7516 | 96.9106 |
|
| 82 |
+
| 0.0012 | 52.26 | 1300 | 2.7878 | 95.3496 |
|
| 83 |
+
| 0.0009 | 56.28 | 1400 | 2.8137 | 97.6260 |
|
| 84 |
+
| 0.0008 | 60.3 | 1500 | 2.8333 | 94.2439 |
|
| 85 |
+
| 0.0007 | 64.32 | 1600 | 2.8514 | 90.1463 |
|
| 86 |
+
| 0.0006 | 68.34 | 1700 | 2.8667 | 95.3821 |
|
| 87 |
+
| 0.0006 | 72.36 | 1800 | 2.8813 | 98.0488 |
|
| 88 |
+
| 0.0005 | 76.38 | 1900 | 2.8932 | 98.8618 |
|
| 89 |
+
| 0.0005 | 80.4 | 2000 | 2.9056 | 98.9268 |
|
| 90 |
+
| 0.0004 | 84.42 | 2100 | 2.9156 | 96.7805 |
|
| 91 |
+
| 0.0004 | 88.44 | 2200 | 2.9251 | 96.7805 |
|
| 92 |
+
| 0.0004 | 92.46 | 2300 | 2.9343 | 97.8211 |
|
| 93 |
+
| 0.0003 | 96.48 | 2400 | 2.9439 | 97.8537 |
|
| 94 |
+
| 0.0003 | 100.5 | 2500 | 2.9516 | 97.1057 |
|
| 95 |
+
| 0.0003 | 104.52 | 2600 | 2.9597 | 98.1138 |
|
| 96 |
+
| 0.0003 | 108.54 | 2700 | 2.9671 | 96.4228 |
|
| 97 |
+
| 0.0003 | 112.56 | 2800 | 2.9733 | 99.1870 |
|
| 98 |
+
| 0.0003 | 116.58 | 2900 | 2.9791 | 102.2764 |
|
| 99 |
+
| 0.0003 | 120.6 | 3000 | 2.9860 | 101.2033 |
|
| 100 |
+
| 0.0002 | 124.62 | 3100 | 2.9903 | 98.9919 |
|
| 101 |
+
| 0.0002 | 128.64 | 3200 | 2.9953 | 98.3415 |
|
| 102 |
+
| 0.0002 | 132.66 | 3300 | 2.9996 | 99.8699 |
|
| 103 |
+
| 0.0002 | 136.68 | 3400 | 3.0034 | 100.1301 |
|
| 104 |
+
| 0.0002 | 140.7 | 3500 | 3.0070 | 98.7317 |
|
| 105 |
+
| 0.0002 | 144.72 | 3600 | 3.0093 | 97.1382 |
|
| 106 |
+
| 0.0002 | 148.74 | 3700 | 3.0118 | 98.3740 |
|
| 107 |
+
| 0.0002 | 152.76 | 3800 | 3.0136 | 96.8130 |
|
| 108 |
+
| 0.0002 | 156.78 | 3900 | 3.0153 | 96.8780 |
|
| 109 |
+
| 0.0002 | 160.8 | 4000 | 3.0153 | 97.2358 |
|
| 110 |
+
|
| 111 |
+
|
| 112 |
+
### Framework versions
|
| 113 |
+
|
| 114 |
+
- Transformers 4.32.0.dev0
|
| 115 |
+
- Pytorch 1.12.1+cu113
|
| 116 |
+
- Datasets 2.14.1
|
| 117 |
+
- Tokenizers 0.13.3
|