Downloads · 30 days
4
10% of all-time downloads
wh-ft-lre5-adm-ga1ba16-reqtm1d
HouraMor/wh-ft-lre5-adm-ga1ba16-reqtm1d
wh-ft-lre5-adm-ga1ba16-reqtm1d is a automatic speech recognition model from HouraMor. 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
40
Public
Parameters
1.5B
12.3 GB on disk
Likes
0
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
- Jun 1, 2025
- Updated
- Jun 1, 2025
- SHA
- 49fc92ff
Base models
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. -->
wh-ft-lre5-adm-ga1ba16-reqtm1d
This model is a fine-tuned version of openai/whisper-large-v3 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9223
- Wer: 0.4763
- Cer: 0.3723
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: 10000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Validation Loss | Wer |
|---|---|---|---|---|---|
| 1.0092 | 0.1994 | 1000 | 0.8200 | 1.0505 | 0.9456 |
| 0.9128 | 0.3989 | 2000 | 0.7055 | 0.9929 | 0.8311 |
| 0.8264 | 0.5983 | 3000 | 0.8438 | 0.9597 | 0.9659 |
| 0.7654 | 0.7978 | 4000 | 0.6011 | 0.9353 | 0.7188 |
| 0.8212 | 0.9972 | 5000 | 1.0073 | 0.9132 | 1.1078 |
| 0.5289 | 1.1966 | 6000 | 0.4207 | 0.9165 | 0.5276 |
| 0.5718 | 1.3961 | 7000 | 0.4991 | 0.9054 | 0.6102 |
| 0.5233 | 1.5955 | 8000 | 0.4271 | 0.8972 | 0.5335 |
| 0.47 | 1.7950 | 9000 | 0.4916 | 0.8876 | 0.6017 |
| 0.3575 | 1.9944 | 10000 | 0.9223 | 0.4763 | 0.3723 |
Framework versions
- Transformers 4.52.3
- Pytorch 2.7.0+cu118
- Datasets 3.5.1
- Tokenizers 0.21.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Jun 1, 2025
- Updated
- Jun 1, 2025