PhoWhisper-medium-vispeech-classifier-v4

This model is a fine-tuned version of vinai/PhoWhisper-medium on the hr16/ViSpeech-Gender-Dialect-Classification dataset. It achieves the following results on the evaluation set:

  • eval_loss: 0.3675
  • eval_accuracy: 0.9131
  • eval_runtime: 574.4362
  • eval_samples_per_second: 5.191
  • eval_steps_per_second: 0.649
  • epoch: 3.0
  • step: 4407

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: 3e-06
  • train_batch_size: 8
  • 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_ratio: 0.1
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.1.2
  • Datasets 2.19.2
  • Tokenizers 0.19.1
Downloads last month
17
Safetensors
Model size
309M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for hr16/PhoWhisper-medium-vispeech-classifier-v1

Finetuned
(1)
this model

Dataset used to train hr16/PhoWhisper-medium-vispeech-classifier-v1