README.md

metadata

library_name: peft
language:
  - en
license: mit
base_model: openai/whisper-large-v3-turbo
tags:
  - wft
  - whisper
  - automatic-speech-recognition
  - audio
  - speech
  - generated_from_trainer
datasets:
  - JacobLinCool/ami-disfluent
model-index:
  - name: whisper-large-v3-turbo-verbatim-3-lora
    results: []

whisper-large-v3-turbo-verbatim-3-lora

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the JacobLinCool/ami-disfluent dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.1459
• eval_wer: 7.7269
• eval_cer: 3.2519
• eval_decode_runtime: 111.0004
• eval_wer_runtime: 0.0705
• eval_cer_runtime: 0.0932
• eval_runtime: 185.4061
• eval_samples_per_second: 10.205
• eval_steps_per_second: 0.324
• epoch: 3.064
• step: 1000

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: 4
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 1000

Framework versions

• PEFT 0.14.0
• Transformers 4.48.0
• Pytorch 2.4.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0