metadata
license: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
datasets:
- common_voice_6_1
metrics:
- wer
model-index:
- name: wav2vec2-large-mms-1b-turkish-colab
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: common_voice_6_1
type: common_voice_6_1
config: tr
split: test
args: tr
metrics:
- name: Wer
type: wer
value: 0.21427841895618424
wav2vec2-large-mms-1b-turkish-colab
This model is a fine-tuned version of facebook/mms-1b-all on the common_voice_6_1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1488
- Wer: 0.2143
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.7357 | 0.46 | 100 | 0.1950 | 0.2664 |
| 0.2959 | 0.92 | 200 | 0.1854 | 0.2666 |
| 0.2657 | 1.38 | 300 | 0.1613 | 0.2279 |
| 0.2565 | 1.83 | 400 | 0.1606 | 0.2266 |
| 0.2564 | 2.29 | 500 | 0.1581 | 0.2259 |
| 0.2418 | 2.75 | 600 | 0.1517 | 0.2186 |
| 0.2559 | 3.21 | 700 | 0.1493 | 0.2150 |
| 0.223 | 3.67 | 800 | 0.1488 | 0.2143 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3