README.md

metadata

license: cc-by-nc-4.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_16_0
base_model: facebook/mms-1b-all
model-index:
  - name: runs
    results: []

runs

This model is a fine-tuned version of facebook/mms-1b-all on the common_voice_16_0 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.1684
• eval_wer: 0.2476
• eval_cer: 0.0432
• eval_runtime: 449.8069
• eval_samples_per_second: 5.985
• eval_steps_per_second: 0.749
• epoch: 10.26
• step: 3000

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.1+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2