Instructions to use espnet/owls_4B_180K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ESPnet
How to use espnet/owls_4B_180K with ESPnet:
from espnet2.bin.asr_inference import Speech2Text model = Speech2Text.from_pretrained( "espnet/owls_4B_180K" ) speech, rate = soundfile.read("speech.wav") text, *_ = model(speech)[0] - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -15,6 +15,8 @@ metrics:
|
|
| 15 |
|
| 16 |
## OWLS: Open Whisper-style Large-scale neural model Suite
|
| 17 |
|
|
|
|
|
|
|
| 18 |
OWLS is a suite of Whisper-style models, designed to help researchers understand the scaling properties of speech models.
|
| 19 |
OWLS models range from 0.25B to 18B parameters, and are trained on up to 360K hours of data.
|
| 20 |
|
|
|
|
| 15 |
|
| 16 |
## OWLS: Open Whisper-style Large-scale neural model Suite
|
| 17 |
|
| 18 |
+
[Paper](https://arxiv.org/abs/2502.10373)
|
| 19 |
+
|
| 20 |
OWLS is a suite of Whisper-style models, designed to help researchers understand the scaling properties of speech models.
|
| 21 |
OWLS models range from 0.25B to 18B parameters, and are trained on up to 360K hours of data.
|
| 22 |
|