Post
The main uses for audio input (at least in our hypothetical string-centric OS) would be saving snippets of it, with or without video, to be played back later or streaming it over the internet via (S)RTP to be played elsewhere.
Then again we might want to incorporate voice commands (traversing links by label), for which we'd need the aid of an NPU. I find Mythic's designs sufficiently weird for my tastes...
Setting that usecase aside it'd mostly be our camera app & SRTP viewer interested.
1/?
Whether supplied via USB or an analog signal our CPU receives a sequence of regularly-sampled numbers. Frequently 16bit 44.1kHz or higher.
Our CPU would use its FPMA coprocessor to aid compressing this data before streaming it to your harddrive or over the internet. I've already discussed details of how this compression works at https://adrian.geek.nz/from-scratch/browser-auditory#audio-recordings .
And we might interleave this into a video, by using container formats (like Ogg, MPEG, USB, or SRTP) to damarcate the substreams.
2/2!
You need to log in to see that page.
Install bonfire.cafe
Get the full app experience
