Dadangdut33 / Speech-Translate

A realtime speech transcription and translation application using Whisper OpenAI and free translation API. Interface made using Tkinter. Code written fully in Python.
MIT License
423 stars 55 forks source link

[REQ] Refinement feature should take .srt files, but only takes .json #78

Open Benko111 opened 3 months ago

Benko111 commented 3 months ago

Is your feature request related to a problem? Please describe. I love align + refine features of your software, it's just great and I've got 600 transcripts ready to be processed. However, the tooltip for refinement suggests to pick either .srt or .json file, but then the filepicker would only take .json. Would be great if this could be fixed so I can use .srt files as well for refinement, because after that I'm feeding the whole thing into another tool to create a specific datetime format (building syncronized transcripts for my podcasts) and this tool only takes .srt files.

Improvement for Alignment feature When using alingment the filepicker for "alignment file" will always default back to .json. So when adding a dozen txt-files for alignment I every time have to revert back the filepicker to look for txt-files. Maybe this could be changed so the filepicker remembers last file format. It's a tiny nuisance, but very annoying impact when adding 20, 30, 40 files at once. Thank you.