Downloads · 30 days
9
2% of all-time downloads
emre/wav2vec2-xls-r-300m-Tr-med-CommonVoice8-Tr-med-CommonVoice8
wav2vec2-xls-r-300m-Tr-med-CommonVoice8-Tr-med-CommonVoice8 is a automatic speech recognition model from emre. 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
9
2% of all-time downloads
All-time downloads
368
Public
Repo size
3.8 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 emre/wav2vec2-xls-r-300m-Tr-med-CommonVoice8 on the common_voice 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 |
|---|---|---|---|---|
| 2.0402 | 0.67 | 500 | 0.3354 | 0.5681 |
| 0.7265 | 1.33 | 1000 | 0.3181 | 0.5444 |
| 0.6858 | 2.0 | 1500 | 0.3044 | 0.5322 |
| 0.6537 | 2.66 | 2000 | 0.2911 | 0.5217 |
| 0.6337 | 3.33 | 2500 | 0.2874 | 0.5164 |
| 0.6111 | 3.99 | 3000 | 0.2758 | 0.5059 |
| 0.5815 | 4.66 | 3500 | 0.2708 | 0.5010 |