Feature Request: Custom Save Paths, Language Identification, and Real-Time Recording Features

[sebvannistel]

Hello, first of all, thank you for developing such a useful tool! I've been using it for a while now and it's been incredibly helpful. I have several feature suggestions that I believe could make the tool even more powerful and user-friendly.

Custom Save Paths for Transcripts: Currently, it's not clear where the transcribed text files are being saved. It would be great to have an option in the preferences to set a custom path for saving these files.

Language Identification in Transcripts: It would be incredibly useful to have the transcript also display what language Whisper has recognized before each line. This could help in multi-language environments and make the transcripts more informative.

Real-Time Recording to TXT Files: More importantly, if there is a possibility to save real-time recordings to txt files in exactly the same way you have already implemented it for audio files. Giving the model, date, language info, etc... This feature is already implemented in docs/docs/preferences.md for audio files but, as far as I understand, not for real-time recordings.

Time and Date in Transcripts: Similar to a WhatsApp conversation, it would be beneficial to have the time and date shown in the transcript.

Handling Multiple Languages in Transcribe Mode: I've noticed that when I'm in transcribe mode, it transcribes English to English but while speaking, when I switch to speaking German, it automatically translates it. It would be cool if it would display the German text as is, or at least give you an option in the preferences to choose how it handles cases like this. Maybe this is an issue with the model and not the app.

Optional Preferences:
All of these features could be something that can be enabled or disabled in the preferences, giving users the flexibility to choose based on their specific needs.

Why These Features?
Custom Save Paths: This would give users more control over their file system and make it easier to organize transcripts.

Language Identification: This would be particularly useful for users who are dealing with multi-language audio and would like to have a more comprehensive transcript.

Real-Time Recording to TXT Files: This would make real-time transcriptions more accessible and easier to review later.

Time and Date in Transcripts: This would make it easier to reference specific parts of a conversation or recording.

Handling Multiple Languages: This would provide a more accurate and user-friendly experience when dealing with multi-language audio.

I believe these features would be a valuable addition to the tool and would love to hear the community's thoughts on this. Thank you for considering my suggestions!