File size: 2,056 Bytes
0e28b95 b6b8dfa 0e28b95 b6b8dfa 0e28b95 b6b8dfa 0e28b95 b6b8dfa |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
---
library_name: transformers
base_model: openai/whisper-large-v3-turbo
tags:
- generated_from_trainer
datasets:
- kojo-george/asanti-twi-tts
metrics:
- wer
model-index:
- name: Whisper ASR Asanti Twi
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: kojo-george/asanti-twi-tts
type: asanti-twi-dataset
args: 'config: hi, split: test'
metrics:
- name: Wer
type: wer
value: 18.398768283294842
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Whisper ASR Asanti Twi
This model is a fine-tuned version of [openai/whisper-turbo](https://huggingface.co/openai/whisper-turbo) on the kojo-george/asanti-twi-tts dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2205
- Wer: 18.3988
## 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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use 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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:-------:|
| 0.226 | 0.5666 | 1000 | 0.3430 | 25.6197 |
| 0.1438 | 1.1331 | 2000 | 0.2737 | 20.8776 |
| 0.1277 | 1.6997 | 3000 | 0.2353 | 18.9530 |
| 0.083 | 2.2663 | 4000 | 0.2205 | 18.3988 |
### Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3 |