Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- it
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- hf-asr-leaderboard
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- mozilla-foundation/common_voice_11_0
|
| 10 |
+
metrics:
|
| 11 |
+
- wer
|
| 12 |
+
model-index:
|
| 13 |
+
- name: Whisper Small Italian
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
name: Automatic Speech Recognition
|
| 17 |
+
type: automatic-speech-recognition
|
| 18 |
+
dataset:
|
| 19 |
+
name: Common Voice 11.0
|
| 20 |
+
type: mozilla-foundation/common_voice_11_0
|
| 21 |
+
args: 'config: it, split: test'
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Wer
|
| 24 |
+
type: wer
|
| 25 |
+
value: 17.391605006569392
|
| 26 |
+
---
|
| 27 |
+
# Whisper Small Italian
|
| 28 |
+
|
| 29 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the Common Voice 11.0 dataset.
|
| 30 |
+
It achieves the following results on the evaluation set:
|
| 31 |
+
- Loss: 0.1185
|
| 32 |
+
- Wer: 17.3916
|
| 33 |
+
|
| 34 |
+
## Model description
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Intended uses & limitations
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Training and evaluation data
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training procedure
|
| 47 |
+
|
| 48 |
+
### Training hyperparameters
|
| 49 |
+
|
| 50 |
+
The following hyperparameters were used during training:
|
| 51 |
+
- learning_rate: 1e-05
|
| 52 |
+
- train_batch_size: 16
|
| 53 |
+
- eval_batch_size: 8
|
| 54 |
+
<!-- - seed: 42 -->
|
| 55 |
+
- gradient_accumulation_steps: 1
|
| 56 |
+
- total_train_batch_size: 16
|
| 57 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 58 |
+
- lr_scheduler_type: linear
|
| 59 |
+
- lr_scheduler_warmup_steps: 500
|
| 60 |
+
- training_steps: 954
|
| 61 |
+
<!-- - mixed_precision_training: Native AMP -->
|
| 62 |
+
|
| 63 |
+
### Training results
|
| 64 |
+
|
| 65 |
+
| Training Loss | Step | Validation Loss | Wer |
|
| 66 |
+
|:-------------:|:----:|:---------------:|:-------:|
|
| 67 |
+
| 1.4744 | 100 | 1.1852 | 117.6059 |
|
| 68 |
+
| 0.7241 | 200 | 0.7452 | 79.7386 |
|
| 69 |
+
| 0.3321 | 300 | 0.3215 | 21.0497 |
|
| 70 |
+
| 0.2930 | 400 | 0.3030 | 20.2129 |
|
| 71 |
+
| 0.2698 | 500 | 0.2982 | 19.7635 |
|
| 72 |
+
| 0.2453 | 600 | 0.2898 | 19.0097 |
|
| 73 |
+
| 0.2338 | 700 | 0.2768 | 18.7054 |
|
| 74 |
+
| 0.2402 | 800 | 0.2646 | 18.2214 |
|
| 75 |
+
| 0.2340 | 900 | 0.2581 | 17.3916 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.25.0.dev0
|
| 81 |
+
- Pytorch 1.12.1+cu113
|
| 82 |
+
- Datasets 2.7.1
|
| 83 |
+
- Tokenizers 0.13.2
|