Downloads · 30 days
6
3% of all-time downloads
dennohpeter/wav2vec2-xls-r-300m-50e-sw-asr
wav2vec2-xls-r-300m-50e-sw-asr is a automatic speech recognition model from dennohpeter. 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
6
3% of all-time downloads
All-time downloads
204
Public
Parameters
315M
30.3 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 facebook/wav2vec2-xls-r-300m on an Mozilla Common Voice 24_0 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 |
|---|---|---|---|---|
| 3.173600 | 0.0690 | 1600 | 0.156657 | 0.2915 |
| 0.391000 | 0.1380 | 3200 | 0.075415 | 0.2077 |
| 0.290000 | 0.2070 | 4800 | 0.058343 | 0.1862 |
| 0.249500 | 0.2760 | 6400 | 0.051531 | 0.1750 |
| 0.223800 | 0.3450 | 8000 | 0.048635 | 0.1719 |
| 0.207800 | 0.4140 | 9600 | 0.046165 | 0.1665 |
| 0.194500 | 0.4830 | 11200 | 0.042730 | 0.1627 |
| 0.182100 | 0.5520 | 12800 | 0.043094 | 0.1630 |
| 0.172300 | 0.6210 | 14400 | 0.040012 | 0.1587 |
| 0.163100 | 0.6900 | 16000 | 0.039247 | 0.1565 |
| 0.155100 | 0.7590 | 17600 | 0.041266 | 0.1580 |
| 0.149100 | 0.8280 | 19200 | 0.043892 | 0.1589 |
| 0.143600 | 0.8970 | 20800 | 0.039167 | 0.1536 |
| 0.135400 | 0.9660 | 22400 | 0.037973 | 0.1520 |
| 0.129900 | 1.0350 | 24000 | 0.038127 | 0.1533 |
| 0.125200 | 1.1040 | 25600 | 0.038621 | 0.1526 |
| 0.119400 | 1.1730 | 27200 | 0.037926 | 0.1517 |
| 0.113400 | 1.2420 | 28800 | 0.037436 | 0.1509 |
| 0.108700 | 1.3110 | 30400 | 0.037197 | 0.1506 |
| 0.105300 | 1.3800 | 32000 | 0.037065 | 0.1511 |