Though video (with an audio/voice track) need subtitles (perhaps a transcript too, especially if it’s long), audio recording alone can be even more inaccessible if not supplemented. To this end, I’m investigating how transcripts can be generated (if you have not made an audio recording from a pre-written script or notes).
Within Google Docs there is a Voice Typing tool (under the Tools menu). Once selected, you can either play audio recording (through the system audio) or speak directly into a microphone, and see the text appear on the screen. It doesn’t attempt to add punctuation, so you would have to add that later.
I have tried this for dictation, and creating a transcript from the audio tracks recorded by others, but so far only those with US or English accents without a pronounced regional accent. My experience so far as been positive; I am neither a fast nor accurate typist and this is an improvement on anything I can achieve already. But the success of this is obviously going to be dependent on its ability to interpret accents, and I would be interested to know how others find this.