update model card README.md
Browse files
README.md
CHANGED
|
@@ -1,39 +1,24 @@
|
|
| 1 |
---
|
| 2 |
-
language:
|
| 3 |
-
- ur
|
| 4 |
license: apache-2.0
|
|
|
|
| 5 |
tags:
|
| 6 |
-
- hf-asr-leaderboard
|
| 7 |
- generated_from_trainer
|
| 8 |
-
datasets:
|
| 9 |
-
- mozilla-foundation/common_voice_7_0
|
| 10 |
metrics:
|
| 11 |
- wer
|
| 12 |
model-index:
|
| 13 |
-
- name:
|
| 14 |
-
results:
|
| 15 |
-
- task:
|
| 16 |
-
name: Automatic Speech Recognition
|
| 17 |
-
type: automatic-speech-recognition
|
| 18 |
-
dataset:
|
| 19 |
-
name: Common Voice 7.0
|
| 20 |
-
type: mozilla-foundation/common_voice_7_0
|
| 21 |
-
args: 'config: ur, split: test'
|
| 22 |
-
metrics:
|
| 23 |
-
- name: Wer
|
| 24 |
-
type: wer
|
| 25 |
-
value: 47.34088927637315
|
| 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 |
-
#
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Wer:
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
@@ -59,21 +44,24 @@ The following hyperparameters were used during training:
|
|
| 59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
- lr_scheduler_type: linear
|
| 61 |
- lr_scheduler_warmup_steps: 100
|
| 62 |
-
- training_steps:
|
| 63 |
-
- mixed_precision_training: Native AMP
|
| 64 |
|
| 65 |
### Training results
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 68 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 69 |
-
|
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
|
| 73 |
|
| 74 |
### Framework versions
|
| 75 |
|
| 76 |
-
- Transformers 4.
|
| 77 |
-
- Pytorch
|
| 78 |
-
- Datasets 2.
|
| 79 |
-
- Tokenizers 0.13.
|
|
|
|
| 1 |
---
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
+
base_model: openai/whisper-small
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- wer
|
| 8 |
model-index:
|
| 9 |
+
- name: whisper-small-ur
|
| 10 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
|
| 16 |
+
# whisper-small-ur
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0029
|
| 21 |
+
- Wer: 76.1310
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_steps: 100
|
| 47 |
+
- training_steps: 700
|
|
|
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 1.1678 | 2.27 | 100 | 1.0347 | 110.8789 |
|
| 54 |
+
| 0.3171 | 4.55 | 200 | 0.2896 | 54.6101 |
|
| 55 |
+
| 0.1408 | 6.82 | 300 | 0.0993 | 98.5782 |
|
| 56 |
+
| 0.0447 | 9.09 | 400 | 0.0261 | 38.4102 |
|
| 57 |
+
| 0.0101 | 11.36 | 500 | 0.0083 | 74.2783 |
|
| 58 |
+
| 0.004 | 13.64 | 600 | 0.0035 | 72.9642 |
|
| 59 |
+
| 0.0031 | 15.91 | 700 | 0.0029 | 76.1310 |
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
| 63 |
|
| 64 |
+
- Transformers 4.32.0.dev0
|
| 65 |
+
- Pytorch 2.0.1+cu118
|
| 66 |
+
- Datasets 2.14.1
|
| 67 |
+
- Tokenizers 0.13.3
|