End of training
Browse files
README.md
CHANGED
|
@@ -10,7 +10,7 @@ datasets:
|
|
| 10 |
metrics:
|
| 11 |
- wer
|
| 12 |
model-index:
|
| 13 |
-
- name:
|
| 14 |
results:
|
| 15 |
- task:
|
| 16 |
name: Automatic Speech Recognition
|
|
@@ -20,22 +20,22 @@ model-index:
|
|
| 20 |
type: mozilla-foundation/common_voice_17_0
|
| 21 |
config: hu
|
| 22 |
split: None
|
| 23 |
-
args:
|
| 24 |
metrics:
|
| 25 |
- name: Wer
|
| 26 |
type: wer
|
| 27 |
-
value:
|
| 28 |
---
|
| 29 |
|
| 30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 31 |
should probably proofread and complete it, then remove this comment. -->
|
| 32 |
|
| 33 |
-
#
|
| 34 |
|
| 35 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
-
- Loss: 0.
|
| 38 |
-
- Wer:
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
@@ -55,26 +55,26 @@ More information needed
|
|
| 55 |
|
| 56 |
The following hyperparameters were used during training:
|
| 57 |
- learning_rate: 1e-05
|
| 58 |
-
- train_batch_size:
|
| 59 |
-
- eval_batch_size:
|
| 60 |
- seed: 42
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
-
- lr_scheduler_warmup_steps:
|
| 64 |
-
- training_steps:
|
| 65 |
- mixed_precision_training: Native AMP
|
| 66 |
|
| 67 |
### Training results
|
| 68 |
|
| 69 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 70 |
|:-------------:|:------:|:----:|:---------------:|:-------:|
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
| 76 |
|
| 77 |
- Transformers 4.40.1
|
| 78 |
- Pytorch 2.2.1+cu121
|
| 79 |
-
- Datasets 2.19.
|
| 80 |
- Tokenizers 0.19.1
|
|
|
|
| 10 |
metrics:
|
| 11 |
- wer
|
| 12 |
model-index:
|
| 13 |
+
- name: whisper-small-Hu-2000-Lyhourt
|
| 14 |
results:
|
| 15 |
- task:
|
| 16 |
name: Automatic Speech Recognition
|
|
|
|
| 20 |
type: mozilla-foundation/common_voice_17_0
|
| 21 |
config: hu
|
| 22 |
split: None
|
| 23 |
+
args: hu
|
| 24 |
metrics:
|
| 25 |
- name: Wer
|
| 26 |
type: wer
|
| 27 |
+
value: 24.247306053771524
|
| 28 |
---
|
| 29 |
|
| 30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 31 |
should probably proofread and complete it, then remove this comment. -->
|
| 32 |
|
| 33 |
+
# whisper-small-Hu-2000-Lyhourt
|
| 34 |
|
| 35 |
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 17.0 dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
+
- Loss: 0.2443
|
| 38 |
+
- Wer: 24.2473
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
|
|
| 55 |
|
| 56 |
The following hyperparameters were used during training:
|
| 57 |
- learning_rate: 1e-05
|
| 58 |
+
- train_batch_size: 64
|
| 59 |
+
- eval_batch_size: 32
|
| 60 |
- seed: 42
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
+
- lr_scheduler_warmup_steps: 50
|
| 64 |
+
- training_steps: 500
|
| 65 |
- mixed_precision_training: Native AMP
|
| 66 |
|
| 67 |
### Training results
|
| 68 |
|
| 69 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 70 |
|:-------------:|:------:|:----:|:---------------:|:-------:|
|
| 71 |
+
| 0.1546 | 0.3298 | 250 | 0.2641 | 25.3667 |
|
| 72 |
+
| 0.2255 | 0.6596 | 500 | 0.2443 | 24.2473 |
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
| 76 |
|
| 77 |
- Transformers 4.40.1
|
| 78 |
- Pytorch 2.2.1+cu121
|
| 79 |
+
- Datasets 2.19.1
|
| 80 |
- Tokenizers 0.19.1
|