Downloads · 30 days
8
35% of all-time downloads
nkkbr/whisper-large-v3-zatoichi-ja-EX-6
whisper-large-v3-zatoichi-ja-EX-6 is a automatic speech recognition model from nkkbr. 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. --
Downloads · 30 days
8
35% of all-time downloads
All-time downloads
23
Public
Parameters
1.5B
30.9 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors6.2 GB · 100%
From the Hugging Face model README
This model is a fine-tuned version of openai/whisper-large-v3 on the None dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.1075 | 0.1852 | 5 | 0.9643 | 110.7085 |
| 0.6924 | 0.3704 | 10 | 0.7488 | 110.8696 |
| 0.4785 | 0.5556 | 15 | 0.6641 | 70.5314 |
| 0.6296 | 0.7407 | 20 | 0.6189 | 75.2818 |
| 0.5476 | 0.9259 | 25 | 0.5828 | 71.1755 |
| 0.372 | 1.1111 | 30 | 0.5699 | 76.9726 |
| 0.2897 | 1.2963 | 35 | 0.5694 | 68.3575 |
| 0.4449 | 1.4815 | 40 | 0.5524 | 64.2512 |
| 0.3885 | 1.6667 | 45 | 0.5445 | 70.0483 |
| 0.2745 | 1.8519 | 50 | 0.5357 | 70.0483 |
| 0.2569 | 2.0370 | 55 | 0.5385 | 66.1031 |
| 0.2397 | 2.2222 | 60 | 0.5409 | 68.3575 |
| 0.2643 | 2.4074 | 65 | 0.5520 | 63.2850 |
| 0.2122 | 2.5926 | 70 | 0.5560 | 70.2093 |
| 0.2569 | 2.7778 | 75 | 0.5496 | 64.8148 |
| 0.2166 | 2.9630 | 80 | 0.5483 | 64.4122 |
| 0.146 | 3.1481 | 85 | 0.5503 | 70.2899 |
| 0.17 | 3.3333 | 90 | 0.5584 | 69.4042 |
| 0.1662 | 3.5185 | 95 | 0.5671 | 66.6667 |
| 0.1667 | 3.7037 | 100 | 0.5667 | 63.2850 |
| 0.1508 | 3.8889 | 105 | 0.5675 | 62.9630 |