Downloads · 30 days
19
3% of all-time downloads
DrishtiSharma/wav2vec2-large-xls-r-300m-hi-cv8-b2
wav2vec2-large-xls-r-300m-hi-cv8-b2 is a automatic speech recognition model from DrishtiSharma. 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
19
3% of all-time downloads
All-time downloads
613
Public
Repo size
5 GB
Likes
0
Public
Click a slice to open those files.
.bin1.3 GB · 100%
From the Hugging Face model README
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - HI dataset. It achieves the following results on the evaluation set:
python eval.py --model_id DrishtiSharma/wav2vec2-large-xls-r-300m-hi-cv8-b2 --dataset mozilla-foundation/common_voice_8_0 --config hi --split test --log_outputs
Hindi language isn't available in speech-recognition-community-v2/dev_data
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 9.6226 | 1.04 | 200 | 3.8855 | 1.0 |
| 3.4678 | 2.07 | 400 | 3.4283 | 1.0 |
| 2.3668 | 3.11 | 600 | 1.0743 | 0.7175 |
| 0.7308 | 4.15 | 800 | 0.7663 | 0.5498 |
| 0.4985 | 5.18 | 1000 | 0.6957 | 0.5001 |
| 0.3817 | 6.22 | 1200 | 0.6932 | 0.4866 |
| 0.3281 | 7.25 | 1400 | 0.7034 | 0.4983 |
| 0.2752 | 8.29 | 1600 | 0.6588 | 0.4606 |
| 0.2475 | 9.33 | 1800 | 0.6514 | 0.4328 |
| 0.219 | 10.36 | 2000 | 0.6396 | 0.4176 |
| 0.2036 | 11.4 | 2200 | 0.6867 | 0.4162 |
| 0.1793 | 12.44 | 2400 | 0.6943 | 0.4196 |
| 0.1724 | 13.47 | 2600 | 0.6862 | 0.4260 |
| 0.1554 | 14.51 | 2800 | 0.7615 | 0.4222 |
| 0.151 | 15.54 | 3000 | 0.7058 | 0.4110 |
| 0.1335 | 16.58 | 3200 | 0.7172 | 0.3986 |
| 0.1326 | 17.62 | 3400 | 0.7182 | 0.3923 |
| 0.1225 | 18.65 | 3600 | 0.6995 | 0.3910 |
| 0.1146 | 19.69 | 3800 | 0.7075 | 0.3875 |
| 0.108 | 20.73 | 4000 | 0.7297 | 0.3858 |
| 0.1048 | 21.76 | 4200 | 0.7413 | 0.3850 |
| 0.0979 | 22.8 | 4400 | 0.7452 | 0.3793 |
| 0.0946 | 23.83 | 4600 | 0.7436 | 0.3759 |
| 0.0897 | 24.87 | 4800 | 0.7289 | 0.3754 |
| 0.0854 | 25.91 | 5000 | 0.7271 | 0.3667 |
| 0.0803 | 26.94 | 5200 | 0.7378 | 0.3656 |
| 0.0752 | 27.98 | 5400 | 0.7488 | 0.3680 |
| 0.0718 | 29.02 | 5600 | 0.7185 | 0.3619 |
| 0.0702 | 30.05 | 5800 | 0.7428 | 0.3554 |
| 0.0653 | 31.09 | 6000 | 0.7447 | 0.3559 |
| 0.0638 | 32.12 | 6200 | 0.7327 | 0.3523 |
| 0.058 | 33.16 | 6400 | 0.7339 | 0.3488 |
| 0.0594 | 34.2 | 6600 | 0.7322 | 0.3469 |