Downloads · 30 days
6
10% of all-time downloads
mondhs/liepa3-whisper-small-lt-liepa2_191_minimized_v12
liepa3-whisper-small-lt-liepa2_191_minimized_v12 is a automatic speech recognition model from mondhs. 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
10% of all-time downloads
All-time downloads
61
Public
Parameters
242M
18.4 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors967 MB · 100%
From the Hugging Face model README
This model is a fine-tuned version of openai/whisper-small on the audiofolder 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 |
|---|---|---|---|---|
| 1.2101 | 0.0549 | 500 | 0.6782 | 56.1973 |
| 0.4774 | 0.1097 | 1000 | 0.5338 | 46.8541 |
| 0.3952 | 0.1646 | 1500 | 0.4776 | 41.7287 |
| 0.3437 | 0.2195 | 2000 | 0.4279 | 38.1324 |
| 0.3122 | 0.2743 | 2500 | 0.4019 | 37.1499 |
| 0.2923 | 0.3292 | 3000 | 0.3832 | 35.1695 |
| 0.2735 | 0.3841 | 3500 | 0.3657 | 33.8090 |
| 0.262 | 0.4389 | 4000 | 0.3485 | 32.4110 |
| 0.2431 | 0.4938 | 4500 | 0.3361 | 30.8478 |
| 0.2341 | 0.5487 | 5000 | 0.3245 | 30.0611 |
| 0.2245 | 0.6035 | 5500 | 0.3156 | 29.4566 |
| 0.2202 | 0.6584 | 6000 | 0.3086 | 28.7108 |
| 0.2105 | 0.7133 | 6500 | 0.3015 | 28.0212 |
| 0.2087 | 0.7681 | 7000 | 0.2962 | 27.7334 |
| 0.2001 | 0.8230 | 7500 | 0.2911 | 27.2583 |
| 0.2024 | 0.8779 | 8000 | 0.2870 | 26.9263 |
| 0.1944 | 0.9327 | 8500 | 0.2826 | 26.8548 |
| 0.1948 | 0.9876 | 9000 | 0.2805 | 26.6504 |