Downloads · 30 days
26
3% of all-time downloads
infinitejoy/wav2vec2-large-xls-r-300m-greek
wav2vec2-large-xls-r-300m-greek 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
26
3% of all-time downloads
All-time downloads
874
Public
Repo size
59.3 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 - EL 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 |
|---|---|---|---|---|
| 3.0928 | 4.42 | 500 | 3.0804 | 1.0073 |
| 1.4505 | 8.85 | 1000 | 0.9038 | 0.7330 |
| 1.2207 | 13.27 | 1500 | 0.7375 | 0.6045 |
| 1.0695 | 17.7 | 2000 | 0.7119 | 0.5441 |
| 1.0104 | 22.12 | 2500 | 0.6069 | 0.5296 |
| 0.9299 | 26.55 | 3000 | 0.6168 | 0.5206 |
| 0.8588 | 30.97 | 3500 | 0.6382 | 0.5171 |
| 0.7942 | 35.4 | 4000 | 0.6048 | 0.4988 |
| 0.7808 | 39.82 | 4500 | 0.6730 | 0.5084 |
| 0.743 | 44.25 | 5000 | 0.6749 | 0.5012 |
| 0.6652 | 48.67 | 5500 | 0.6491 | 0.4735 |
| 0.6386 | 53.1 | 6000 | 0.6928 | 0.4954 |
| 0.5945 | 57.52 | 6500 | 0.6359 | 0.4798 |
| 0.5561 | 61.95 | 7000 | 0.6409 | 0.4799 |
| 0.5464 | 66.37 | 7500 | 0.6452 | 0.4691 |
| 0.5119 | 70.8 | 8000 | 0.6376 | 0.4657 |
| 0.474 | 75.22 | 8500 | 0.6541 | 0.4700 |
| 0.45 | 79.65 | 9000 | 0.6374 | 0.4571 |
| 0.4315 | 84.07 | 9500 | 0.6568 | 0.4625 |
| 0.3967 | 88.5 | 10000 | 0.6636 | 0.4605 |
| 0.3937 | 92.92 | 10500 | 0.6537 | 0.4597 |
| 0.3788 | 97.35 | 11000 | 0.6614 | 0.4589 |