❓ Questions / Help / Support

[hungnvk54]

❓ Questions and Help

To Reproduce

Steps to reproduce the behavior:

1. Run get_speech_timestamps(....) with threshold=0.5 with audio in this link: https://drive.google.com/file/d/16rSKOGnm8mBRTpbtQy-kjsTYlBZrijkR/view?usp=sharing

Expected behavior

The model run well in this very clear audio

Environment

1. Silero version 5.1
2. Running with onnx (But had change to jit and the result is same)

[snakers4]