todo #3
Description
- add mic input and output (with auto transcription using speech api) AND GREY TEXT FOR BITS OF SPEECH THAT ARE PROCESSING STILL
- Volume and pitch matching/humanizing
- Add reverse phonemes at some point
- Fix punctuation matching so it actually works again
- add wav and mp3 export and play options in the results
- add result speech editor - and options in this to disable diphones and triphones and change settings like that and add new stuff and delete stuff
- add auto-completion telling you whether or not the word will work
- add multiple input for combined transcripts
- get all broken website things to work
- add a way to use emphasis (CMU dictionary uses 0 = none, 1 = most, 2 = a little)
- convert numbers to words
- add Final Cut type XML export option maybe
- audio effects like normalization or silence removal
- add pitch modification somehow
- add crossfading
- add phone alternatives to avoid OOV
- add automatic click/pop removal
- add a way to set the audio transcription microphone
- add a way to set the recording microphone
- midi parsing support
- video playing support
After it works properly:
- maybe do something where you could have a twitch or youtube livestream where you type something and it says it in another voice and the voices rotate every now and then