Commit
•
72ed659
1
Parent(s):
8f173c0
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 Tiny it 6
|
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 |
+
config: it
|
22 |
+
split: test[:10%]
|
23 |
+
args: 'config: it, split: test'
|
24 |
+
metrics:
|
25 |
+
- name: Wer
|
26 |
+
type: wer
|
27 |
+
value: 97.56655574043262)
|
28 |
+
---
|
29 |
+
# Whisper Tiny it 6
|
30 |
+
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Common Voice 11.0 dataset.
|
31 |
+
It achieves the following results on the evaluation set:
|
32 |
+
- Loss: 2.137834
|
33 |
+
- Wer: 97.566556
|
34 |
+
|
35 |
+
## Model description
|
36 |
+
|
37 |
+
This model is the openai whisper small transformer adapted for Italian audio to text transcription.
|
38 |
+
As part of the hyperparameter tuning process weight decay set to 0.1, attention dropout, encoder dropout and decoder dropout have been set to 0.1,
|
39 |
+
the learning rate has been set to 1e-5, the number of decoder attention heads and encoder attention heads have been set to 8
|
40 |
+
however, it did not improved the performance on the evaluation set.
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
The model is available through its [HuggingFace web app](https://huggingface.co/spaces/GIanlucaRub/whisper-it)
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
Data used for training is the initial 10% of train and validation of [Italian Common Voice](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0/viewer/it/train) 11.0 from Mozilla Foundation.
|
49 |
+
The dataset used for evaluation is the initial 10% of test of Italian Common Voice.
|
50 |
+
|
51 |
+
|
52 |
+
## Training procedure
|
53 |
+
|
54 |
+
After loading the pre trained model, it has been trained on the dataset.
|
55 |
+
|
56 |
+
### Training hyperparameters
|
57 |
+
|
58 |
+
The following hyperparameters were used during training:
|
59 |
+
- learning_rate: 1e-05
|
60 |
+
- train_batch_size: 16
|
61 |
+
- eval_batch_size: 8
|
62 |
+
- seed: 42
|
63 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
64 |
+
- lr_scheduler_type: linear
|
65 |
+
- lr_scheduler_warmup_steps: 500
|
66 |
+
- training_steps: 4000
|
67 |
+
- mixed_precision_training: Native AMP
|
68 |
+
### Training results
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
70 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
71 |
+
| 1.7353 | 3.82 | 4000 | 2.1378 | 97.5666 |
|
72 |
+
### Framework versions
|
73 |
+
- Transformers 4.26.0.dev0
|
74 |
+
- Pytorch 1.12.1+cu113
|
75 |
+
- Datasets 2.7.1
|
76 |
+
- Tokenizers 0.13.2
|