README.md

metadata

license: cc-by-nc-4.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_6_1
metrics:
  - wer
base_model: facebook/mms-1b-all
model-index:
  - name: wav2vec2-large-mms-1b-turkish-colab-test
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: common_voice_6_1
          type: common_voice_6_1
          config: tr
          split: test
          args: tr
        metrics:
          - type: wer
            value: 0.22040649576141355
            name: Wer

wav2vec2-large-mms-1b-turkish-colab-test

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: 0.1534
• Wer: 0.2204

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: 32
• 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
4.5985	0.92	100	0.1805	0.2490
0.2839	1.83	200	0.1657	0.2350
0.2662	2.75	300	0.1579	0.2274
0.2413	3.67	400	0.1534	0.2204

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.13.0
• Tokenizers 0.13.3