Downloads · 30 days
2
10% of all-time downloads
HouraMor/wh-ft-lr5e7-dtstf5-adm-ga1ba16-st15k-pat5-v2-evalstp500
wh-ft-lr5e7-dtstf5-adm-ga1ba16-st15k-pat5-v2-evalstp500 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. --
Downloads · 30 days
2
10% of all-time downloads
All-time downloads
21
Public
Parameters
1.5B
111 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 an unknown 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 | Cer |
|---|---|---|---|---|---|
| 1.7517 | 0.3047 | 500 | 1.7105 | 0.2620 | 0.1986 |
| 0.6885 | 0.6094 | 1000 | 0.6924 | 0.2530 | 0.1932 |
| 0.6446 | 0.9141 | 1500 | 0.6360 | 0.2534 | 0.1867 |
| 0.5855 | 1.2188 | 2000 | 0.6108 | 0.2495 | 0.1842 |
| 0.5408 | 1.5235 | 2500 | 0.5976 | 0.2480 | 0.1878 |
| 0.5859 | 1.8282 | 3000 | 0.5841 | 0.2386 | 0.1786 |
| 0.5058 | 2.1328 | 3500 | 0.5778 | 0.2372 | 0.1771 |
| 0.4703 | 2.4375 | 4000 | 0.5725 | 0.2412 | 0.1859 |
| 0.4945 | 2.7422 | 4500 | 0.5662 | 0.2386 | 0.1788 |
| 0.4734 | 3.0469 | 5000 | 0.5633 | 0.2418 | 0.1873 |
| 0.4436 | 3.3516 | 5500 | 0.5632 | 0.2382 | 0.1808 |
| 0.4584 | 3.6563 | 6000 | 0.5619 | 0.2364 | 0.1843 |
| 0.4535 | 3.9610 | 6500 | 0.5591 | 0.2384 | 0.1806 |
| 0.4212 | 4.2657 | 7000 | 0.5641 | 0.2428 | 0.1867 |
| 0.3701 | 4.5704 | 7500 | 0.5634 | 0.2395 | 0.1861 |
| 0.4144 | 4.8751 | 8000 | 0.5615 | 0.2394 | 0.1854 |
| 0.3616 | 5.1798 | 8500 | 0.5690 | 0.2410 | 0.1838 |
| 0.3742 | 5.4845 | 9000 | 0.5701 | 0.2403 | 0.1856 |