Downloads · 30 days
8
20% of all-time downloads
dianavdavidson/indicwav2vec-hindi-iv_hindi_only-62143-normalized-2e-5-epochs-100-FT
indicwav2vec-hindi-iv_hindi_only-62143-normalized-2e-5-epochs-100-FT is a automatic speech recognition model from dianavdavidson. 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
20% of all-time downloads
All-time downloads
41
Public
Parameters
316M
1.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors1.3 GB · 100%
From the Hugging Face model README
This model is a fine-tuned version of ai4bharat/indicwav2vec-hindi 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 | Global Wer |
|---|---|---|---|---|
| 22.2447 | 1.0 | 540 | 3.2673 | 100.0 |
| 4.3766 | 2.0 | 1080 | 0.9075 | 40.5162 |
| 2.1337 | 3.0 | 1620 | 0.6814 | 30.7721 |
| 1.7836 | 4.0 | 2160 | 0.6033 | 27.1257 |
| 1.6091 | 5.0 | 2700 | 0.5624 | 25.5890 |
| 1.5184 | 6.0 | 3240 | 0.5463 | 24.5511 |
| 1.4624 | 7.0 | 3780 | 0.5251 | 24.0953 |
| 1.4131 | 8.0 | 4320 | 0.5229 | 23.5964 |
| 1.3978 | 9.0 | 4860 | 0.5425 | 24.0799 |
| 1.4460 | 10.0 | 5400 | 0.5696 | 24.8375 |
| 1.4575 | 11.0 | 5940 | 0.6076 | 26.7038 |