Implement Auto-Save or Explicit Save Logic for Transcription

[maninthemiddle01]

Is your feature request related to a problem? Please describe.
The current saving mechanism for audio/video transcription is unintuitive and risky. Because exiting via the "Esc" key does not save progress, it appears that the software does not save continuously. This means that if the window is closed accidentally or via a shortcut, all unsaved transcription work is lost.

Describe the solution you'd like
I'd be nice to have a continuous auto-save feature so that every change is captured in real-time. If auto-save is not possible, would it be possible to add a visible "Save" button and ensure that the "Esc" key triggers a save action before closing the window.

[Salomon-Mazatlan]

hi @maninthemiddle01

I have already been working on that file, and I believe that adding a confirmation dialog would be sufficient.

Closing the transcription module typically happens when users open it only to preview or listen to an audio/video file. Improvements related to this workflow are planned as part of a future proposal for the Code Audio/Video module.

[ccbogel]

ok can I leave this update to you @Salomon-Mazatlan
Note that journals.py module does have an autosave feature using a timer

[Salomon-Mazatlan]

ok can I leave this update to you @Salomon-Mazatlan
Note that journals.py module does have an autosave feature using a timer

After the subcode system update, I will submit a proposal for these features. I have been working on several improvements for the view_av and code_av modules.

As a preview, I have added an option in manage_files to import and update transcripts from external files:

I also plan to incorporate the suggested autosave functionality, as well as a transcript importer directly into view_av. For now, I am finishing a number of adjustments to the code_av module:

[ccbogel]

Yes sure. Do not rely on the waveform - as there will be people who will not have or will not be able to install ffmpeg - Windows in particular it is not as easy to install as Linux.

[Salomon-Mazatlan]

Yes sure. Do not rely on the waveform - as there will be people who will not have or will not be able to install ffmpeg - Windows in particular it is not as easy to install as Linux.

Thank you for the observation.

I will run some tests on this, as I implemented a system where the waveform is generated as an image when audio or video files are imported. Coding and navigation are then performed on this visual representation rather than relying on a waveform generated in real time.