Downloads · 30 days
13
3% of all-time downloads
infinitejoy/wav2vec2-large-xls-r-300m-sakha
wav2vec2-large-xls-r-300m-sakha is a automatic speech recognition model from infinitejoy. 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
13
3% of all-time downloads
All-time downloads
483
Public
Repo size
16.4 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_7_0 - SAH 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 |
|---|---|---|---|---|
| 1.8597 | 8.47 | 500 | 0.7731 | 0.7211 |
| 1.2508 | 16.95 | 1000 | 0.5368 | 0.5989 |
| 1.1066 | 25.42 | 1500 | 0.5034 | 0.5533 |
| 1.0064 | 33.9 | 2000 | 0.4686 | 0.5114 |
| 0.9324 | 42.37 | 2500 | 0.4927 | 0.5056 |
| 0.876 | 50.85 | 3000 | 0.4734 | 0.4795 |
| 0.8082 | 59.32 | 3500 | 0.4748 | 0.4799 |
| 0.7604 | 67.8 | 4000 | 0.4949 | 0.4691 |
| 0.7241 | 76.27 | 4500 | 0.5090 | 0.4627 |
| 0.6739 | 84.75 | 5000 | 0.4967 | 0.4452 |
| 0.6447 | 93.22 | 5500 | 0.5071 | 0.4437 |