Hi there, here are some improvements I made this weekend in the FTUE (first-time user experience). First of all, there’s the what’s new section, and then there is guided setup of Gemini inference provider, models, and prompts. In the video, I also show you how to add Whisper for local speech recognition. Please refer to the respective README for how to get the local server running. Ideally, I would like to include the service in the flatpak bundle and automatically start it. Let me know if you’re interested in building & contributing this. Another nice contribution would be the audio waveform visualization similar to how it works on e.g. iOS and macOS.
Sorry for the choppy playback of the talking head in the recording. I tried multiple times with different settings but the problem persisted & I’m too tired now to try anything else. I’m using OBS on macOS for the recordings so far, do you have any recommendations for a better setup?
Cheers,
Matthias
