Does this mean that any software which uses ffmpeg can now add a transcription option? Audacity, Chrome, OBS etc

If they want to support it out-of-the box, they'll still have to embed a model file (roughly 500 MB - 3GB, varying size and quality)