Downloads · 30 days
21
4% of all-time downloads
whisper-large-v3-pt-3000h-4
fsicoli/whisper-large-v3-pt-3000h-4
whisper-large-v3-pt-3000h-4 is a automatic speech recognition model from fsicoli. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
535
Public
Parameters
1.5B
24.7 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors6.2 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Aug 17, 2024
- Updated
- Sep 5, 2024
- SHA
- 542335a7
Base models
Datasets
Model card
From the Hugging Face model README
<!-- 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-large-v3-pt-3000h-4
This model is a fine-tuned version of openai/whisper-large-v3 on the fsicoli/common_voice_18_0 pt dataset. It achieves the following results on the evaluation set:
- Loss: 0.1938
- Wer: 0.1081
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-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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: 1000
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0849 | 1.0 | 5529 | 0.1938 | 0.1081 |
| 0.0788 | 2.0 | 11058 | 0.2289 | 0.1061 |
| 0.0183 | 3.0 | 16587 | 0.2809 | 0.1079 |
| 0.0322 | 4.0 | 22116 | 0.3088 | 0.1058 |
| 0.0273 | 5.0 | 27645 | 0.3222 | 0.1038 |
| 0.0204 | 6.0 | 33174 | 0.3532 | 0.1066 |
| 0.0605 | 7.0 | 38703 | 0.3542 | 0.1053 |
| 0.043 | 8.0 | 44232 | 0.3669 | 0.1049 |
| 0.0204 | 9.0 | 49761 | 0.3707 | 0.1036 |
| 0.0159 | 10.0 | 55290 | 0.3697 | 0.1031 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.4.0+cu124
- Datasets 2.18.1.dev0
- Tokenizers 0.19.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Aug 17, 2024
- Updated
- Sep 5, 2024