Gladia (https://www.gladia.io/) transcribes videos or audios into text.

Speaker recognition (“diarization”), translation possible. Export JSON, SRT, VTT and TXT.

Live function not tested.

10 hours of free transcription per month (500 MB files or 2h15 maximum).

Test from the video YouTube: <a href="https://www.youtube.com/watch?v=VMzpDsdvdr0″>”AI, society and education: The challenges for the training of future citizens – Pierre Yves Oudeyer” <a href="https://www.youtube.com/watch?v=VMzpDsdvdr0″>:

Test from mp3 file “The Mirabeau Bridge“