#921 No speech detected. Ensure audio/video contains speech in the selected language.

2804:38a* Posted at: 1 day ago 👁8

No speech detected. Ensure audio/video contains speech in the selected language.

=====

Windows-10-10.0.22631-SP0

version:v3.81

frozen:True

language:en

1 Replies

Floor #1 0.0* Posted at 1 day ago

This article was replied by DeepSeek
The error message "No speech detected. Ensure audio/video contains speech in the selected language" indicates that the speech recognition model could not find any human speech in your audio or video file. This is a common issue with pyVideoTrans. Let's troubleshoot it step by step.

Primary Cause: The most likely reason is that the speech recognition settings are too strict, filtering out the actual speech in your file.

Solution Steps:

1. Check Your Source File: First, confirm that your audio/video file does contain clear human speech in the language you selected ("en" for English). Play the file using a media player to verify this.

2. Adjust Speech Recognition Settings (Most Important Step):

  • In the main software interface, go to the "Speech Recognition" section.
  • Make sure you are using a model like faster-whisper or openai-whisper.
  • Click on the top menu bar: Tools/Options -> Advanced Options.
  • In the window that opens, find the section "faster/openai speech recognition adjustment".
  • Change the following parameters to make the detection more sensitive:
  • Voice Threshold: Set this to 0.3 (lower than the default 0.5 to detect quieter speech).
  • Minimum Speech Duration / ms: Set this to 0 (to include very short speech segments).
  • Silence Split ms: Set this to 500 or 700 (increases the silence period required to split segments, helping to capture longer phrases).
  • Click OK to save these settings.

3. Enable VAD (Voice Activity Detection): In the same "faster/openai speech recognition adjustment" section, ensure the "Enable VAD" checkbox is selected. This helps the model better distinguish between speech and silence.

4. Try a Different Model: If the above doesn't work, switch your Speech Recognition channel to a different model, such as "OpenAI-Whisper (local)" or "Google Speech Recognition" (if you have a proxy configured), and run the task again.

5. Check for Background Noise: If your audio has significant background noise, try enabling the "Noise Reduction" checkbox in the main "Speech Recognition" section.

If you continue to experience this issue after trying these steps, please provide the log file from your software directory videotrans/logs/ (it will be named with the current date, e.g., 20241127.log). This will help in diagnosing the problem further.

Post Your Reply
Open source and free maintenance is not easy. If this project is helpful to you, please consider making a small donation to help the project continue to maintain and update.

Donate: https://ko-fi.com/jianchang512

Trending Questions