Header missing - Transcription skips large sections of audio

[choepke]

Hi!

First of all, I have been using NoScribe for a full semester now to transcribe my Japanese lectures, and it’s been incredibly helpful.

However, recently I started getting the following error message in the terminal whenever I try to transcribe an .mp3 file:

ffmpeg: [mp3 @ 0x...] Estimating duration from bitrate, this may be inaccurate
ffmpeg: [mp3float @ 0x...] Header missing
ffmpeg: [aist#0:0/mp3 @ 0x...] Error submitting packet to decoder: Invalid data found when processing input

When this happens, the transcription finishes without crashing, but only a few minutes at the beginning and end of the file are transcribed, there are large gaps in between that seem to be skipped.

I also tried converting the file manually to .wav format through the terminal before running NoScribe, but I get the same result.
I have been using the exact same setup as before, and these errors just started appearing recently, so I am not sure if it is a problem on my side.

Thank you for your time

System details:
macOS Sequoia 15.5 (24F74)
NoScribe version 0.6

[kaixxx]

It seems that there is a problem with your mp3 recordings. However, I might be able to improve the error handling within noScribe. Could you make a short recording with the same app that you use for your lectures and send it to me so that I can try to reproduce the error? My E-Mail: kai.droege <> hslu.ch

When you tried converting your recording in the terminal; did you use ffmpeg? If so, try adding -err_detect ignore_err before -i so ffmpeg skips bad MP3 frames while decoding. It’s a decoding option and must precede the input.