You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Audio driving servo, auto-talk
- Thread starter grayfx
- Start date
- Status
- Not open for further replies.
Wade Hassler
Member
1) I've done it by generating data to put in PROGMEM by precalculating values corresponding to mouth open degree from my audio files on a PC. This worked well for me since I only had to deal with a fixed amount of audio files and I could fit the data into memory (PROGMEM) and maintain a good enough mouth update rate. If you have many or long files or not to do it on live audio input this obviously won't work.
I even made a tool visualize the output values and for hand-tweaking them but came to conclusion that hand-tweaking was too hard/time consuming and just used the un-edited preprocessing values.
Tool Screenshot were you can see the audio file waveforms along with the values for the mouths (pruple & red bars): http://vvvv.org/blog/preprocessing-and-editing-wip
2) You can probably do it just fine in on the fly with the AUDIO library by just calculating the RMS of the audio and maybe adding some smoothing to not have the output too sensitive.
3) Using external circuits like Wade suggested could also work.
I even made a tool visualize the output values and for hand-tweaking them but came to conclusion that hand-tweaking was too hard/time consuming and just used the un-edited preprocessing values.
Tool Screenshot were you can see the audio file waveforms along with the values for the mouths (pruple & red bars): http://vvvv.org/blog/preprocessing-and-editing-wip
2) You can probably do it just fine in on the fly with the AUDIO library by just calculating the RMS of the audio and maybe adding some smoothing to not have the output too sensitive.
3) Using external circuits like Wade suggested could also work.
Last edited:
- Status
- Not open for further replies.