"list index out of range" is a generic Python indexing error and cannot be pinpointed to one specific cause from this message alone. The exact stage that failed (STT / translation / TTS / alignment / merge) is only recorded in the log.
Please provide the following so the real cause can be identified:
- Open
D:/AI_dub_pyvideotrans-v4.15/logs/ and open the newest .log file (named by date, e.g. 20250622.log). - Copy the last ~30 lines (the full traceback, including the line mentioning which module/file raised the error) and paste it here.
- Also state: which function you were using (Translate Video / Speech-to-Text / Text-to-Speech / Translate SRT), which STT / Translate / TTS channel and model were selected, the spoken language and target language, and whether "Clean Generated (Clear Output)" was checked.
Common triggers for this error, worth checking while you gather the log:
- Audio slice list is empty — the video has no usable audio track, or volume is too low / VAD filtered everything out. Split the audio and confirm human speech is detectable.
- Subtitle/segment count mismatch after AI translation merged or dropped lines (AI translation returning fewer lines than input). Try setting
Menu -> Tools/Options -> Advanced Options -> Translation -> lines per batch for AI translation to 1, and uncheck "Send full SRT". - Corrupted or partially downloaded model files (very common) causing empty recognition results. Delete the related folder under
models/ and re-download. - Special characters / non-ASCII or overly long path or filename for the input video — rename to a simple English name and move to a shallow path like
D:/video/.
Note: since your build is the frozen (packaged) version, please also confirm you have applied the latest v4.15 patch package over it, as several indexing-related bugs were fixed after the initial v4.15 build.
Please check the related docs: