Please implement room correction in pulseaudio

Asked by Henk Poley on 2014-12-02

As a "bribe" I can send a developer who's willing to take a stab at this a UMIK-1 calibration microphone. And I can try to guide the creation of a calibration profile with DRC-FIR.

"Room correction" is basically using a profile of your sound system to lower the volume of all sound frequencies to the point that your speakers are worst at recreating (lowest volume). This effectively emulates a perfect speaker, albeit with a slightly lower sound pressure. It can also try to compensate for echos and reflections in your room. Thanks to the wonders of "convolution" which uses fourier transforms, this all can happen in one filter with one .pcm input file.

Current setups would use pulseaudio on top of either BruteFIR or jack + jconvolver. Of which jconvolver is nicer since it has a lower delay. Both setups are rather brittle. And don't run out of the box after dropping config in place and rebooting. But need you to start the convolution system by hand before (re)starting pulseaudio.

So, all parts of the software already exists. It's mostly knowing the inside of pulseaudio to plumb it all together. A GUI that guides you through the calibration or setup would be a next step, but is not strictly necessary.

Question information

English Edit question
Ubuntu pulseaudio Edit question
No assignee Edit question
Last query:
Last reply:

I suggest you report a bug

Manfred Hampl (m-hampl) said : #2

A better address for such proposals might be

Can you help with this problem?

Provide an answer of your own, or ask Henk Poley for more information if necessary.

To post a message you must log in.