Describe the bug
When transcribing an interview, it seem to work, according to the output. The Log-Message "Transcription Finished" appears and in the summary it is mentioned as Completed. No error is shown.
Even the path where the output file appears is logged with the correct path that was given as input. However, that file does not exist.
Is there a "verbose" Mode of the application, so a bit more log-output can be seen, because without that it is really difficult to see where the problem is.
The transcribing seems to work, despite the older GPU. Its utilization shows that it is used during the transcribing process.
Screenshots
System (please complete the following information):
- OS: Windows 11 with GTX 1660
- On Windows: CUDA Version
- noScribe Version: 0.7.2
- noScribe Settings: Language: both auto and german, model: both fast and precise, mark pause 1s+, Speaker detection: auto, Overlapping Speech: yes, disfluencies: yes, timestamps: no.
Describe the bug
When transcribing an interview, it seem to work, according to the output. The Log-Message "Transcription Finished" appears and in the summary it is mentioned as Completed. No error is shown.
Even the path where the output file appears is logged with the correct path that was given as input. However, that file does not exist.
Is there a "verbose" Mode of the application, so a bit more log-output can be seen, because without that it is really difficult to see where the problem is.
The transcribing seems to work, despite the older GPU. Its utilization shows that it is used during the transcribing process.
Screenshots
System (please complete the following information):