metadata
library_name: transformers
language:
- ro
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
datasets:
- IoanaLiviaPopescu/RealVoiceSynthVoice-800-1-Emil-Neural
metrics:
- wer
model-index:
- name: >-
IoanaLiviaPopescu/IoanaLiviaPopescu/real-data-synth-data-800-1-Emil-Neural-whisper-medium-v1
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: IoanaLiviaPopescu/RealVoiceSynthVoice-800-1-Emil-Neural
type: IoanaLiviaPopescu/RealVoiceSynthVoice-800-1-Emil-Neural
config: default
split: test
args: 'split: validation'
metrics:
- name: Wer
type: wer
value: 9.404388714733543
IoanaLiviaPopescu/IoanaLiviaPopescu/real-data-synth-data-800-1-Emil-Neural-whisper-medium-v1
This model is a fine-tuned version of openai/whisper-medium on the IoanaLiviaPopescu/RealVoiceSynthVoice-800-1-Emil-Neural dataset. It achieves the following results on the evaluation set:
- Loss: 0.2316
- Wer: 9.4044
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1781 | 0.9753 | 37 | 0.2368 | 9.6625 |
| 0.0553 | 1.9753 | 74 | 0.2316 | 9.4044 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1