GIanlucaRub
commited on
Commit
•
bac1f03
1
Parent(s):
72ed659
Update README.md
Browse files
README.md
CHANGED
@@ -10,7 +10,7 @@ datasets:
|
|
10 |
metrics:
|
11 |
- wer
|
12 |
model-index:
|
13 |
-
- name: Whisper Tiny it
|
14 |
results:
|
15 |
- task:
|
16 |
name: Automatic Speech Recognition
|
@@ -26,7 +26,7 @@ model-index:
|
|
26 |
type: wer
|
27 |
value: 97.56655574043262)
|
28 |
---
|
29 |
-
# Whisper Tiny it
|
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
|
@@ -36,7 +36,7 @@ It achieves the following results on the evaluation set:
|
|
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-
|
40 |
however, it did not improved the performance on the evaluation set.
|
41 |
|
42 |
## Intended uses & limitations
|
@@ -56,7 +56,7 @@ After loading the pre trained model, it has been trained on the dataset.
|
|
56 |
### Training hyperparameters
|
57 |
|
58 |
The following hyperparameters were used during training:
|
59 |
-
- learning_rate: 1e-
|
60 |
- train_batch_size: 16
|
61 |
- eval_batch_size: 8
|
62 |
- seed: 42
|
|
|
10 |
metrics:
|
11 |
- wer
|
12 |
model-index:
|
13 |
+
- name: Whisper Tiny it 7
|
14 |
results:
|
15 |
- task:
|
16 |
name: Automatic Speech Recognition
|
|
|
26 |
type: wer
|
27 |
value: 97.56655574043262)
|
28 |
---
|
29 |
+
# Whisper Tiny it 7
|
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
|
|
|
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-6, 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
|
|
|
56 |
### Training hyperparameters
|
57 |
|
58 |
The following hyperparameters were used during training:
|
59 |
+
- learning_rate: 1e-06
|
60 |
- train_batch_size: 16
|
61 |
- eval_batch_size: 8
|
62 |
- seed: 42
|