Downloads · 30 days
26
4% of all-time downloads
NX2411/wav2vec2-large-xlsr-korean-demo-colab-2
wav2vec2-large-xlsr-korean-demo-colab-2 is a automatic speech recognition model from NX2411. 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
4% of all-time downloads
All-time downloads
726
Public
Repo size
6.3 GB
Likes
0
Public
Click a slice to open those files.
.bin1.3 GB · 99%
From the Hugging Face model README
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None 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 |
|---|---|---|---|---|
| 4.7387 | 2.12 | 400 | 3.1791 | 1.0 |
| 1.3766 | 4.23 | 800 | 0.4876 | 0.5264 |
| 0.476 | 6.35 | 1200 | 0.2955 | 0.3648 |
| 0.3209 | 8.46 | 1600 | 0.2926 | 0.3473 |
| 0.2591 | 10.58 | 2000 | 0.2723 | 0.3094 |
| 0.2055 | 12.7 | 2400 | 0.2746 | 0.3027 |
| 0.1802 | 14.81 | 2800 | 0.2672 | 0.2976 |
| 0.1552 | 16.93 | 3200 | 0.2822 | 0.2807 |
| 0.1413 | 19.05 | 3600 | 0.2652 | 0.2856 |
| 0.1232 | 21.16 | 4000 | 0.2631 | 0.2655 |
| 0.1146 | 23.28 | 4400 | 0.2561 | 0.2574 |
| 0.1086 | 25.4 | 4800 | 0.2461 | 0.2527 |
| 0.0944 | 27.51 | 5200 | 0.2521 | 0.2535 |
| 0.0881 | 29.63 | 5600 | 0.2481 | 0.2480 |