1276928 - MediaRecorder: clarify what happens when recording a stream with multiple tracks with different audio sampling rates

Status: NEW

(Core :: Audio/Video: Recording, defect, P4)

Description

[Soledad Penades [:sole] [:spenades]]

What should happen? Should the output be encoded at the maximum sampling rate? Can various tracks with different sampling rates coexist? Will they be upsampled to all match the highest rate?

This isn't defined in the MediaRecorder spec either, but it's getting developers confused.

Linking to the bug that already exists, for tracking purposes.

Comment 1

[Byron Campen [:bwc]]

Does this prioritization seem sane?

Comment 2

[Paul Adenot (:padenot)]

(In reply to Soledad Penades [:sole] [:spenades] from comment #0)
> What should happen? Should the output be encoded at the maximum sampling
> rate? Can various tracks with different sampling rates coexist? Will they be
> upsampled to all match the highest 

What would be a setup where this happens ?

For now in Gecko (and I'm just talking about the current setup to give some sense of perspective), it's all resampled to the "native sample rate" of the device the audio is going to (usually, the default audio output of the system), even if no audio is going out of the speakers.

This is going to change, because we'll be able to specify the sample rate of AudioContexts, and then as you mention we'll really have to spec what should happen.

Comment 3

[Bulk Bug Changes for mreavy's org]

Mass change P3->P4 to align with new Mozilla triage process.