Ray-CubeAce
Second Officer
Offline
|
TheCameraLens Posted at 9-30 13:41
I had a feeling that you knew what you were talking about and that you knew the mic was OK. I thought I would just ask the question as it is all too easy to get caught up in one train of thought. I know I have been guilty of that many times. I hope that you find a resolution to the issue.
The only resolution anyone has found here to date is to use an external recorder and then sync the output in post production. A pity because in all other aspects the sound from the Osmo Pocket is reasonably high quality and the auto leveling function and clipping overload function has been second to none.
My guess, and it's only a guess is that initially the second mic and it's positioning on the Osmo Pocket was only intended for wind noise cancellation which was not only not very effective but people complained the resulting sound was unacceptable. DJI was quick to respond to that and something was done to a firmware update that improved things. This was all done before the release of the audio adapter. The resulting output was not true stereo either but we put that down to the positioning of the mics and hoped the audio adapter would rectify that. We felt that using an external mic with sufficient wind protection would be a good solution, but it turned out what we got was not the weird sounding output that the Osmo Pocket itself produced but what seemed like an adapter that was very fussy about which microphones worked best with it and regardless of wiring configuration and what to anyone with two functioning ears sounds completely mono. Two channels or not. One channel only (either side, doesn't matter) produces no sound at all and both channels wired to independent mic capsules gives a non spacial sound-stage.
It was only after the release of the adapter that we found that problem.
If you can inspect the video info it reports two independent channels for audio within the aac. stream at 192kb/s @ 48kHz. with an independent frame rate of 46.875 FPS regardless of the frame rate of the video, but that's not uncommon.
However, put the sound file into an audio editor capable of splitting up the left and right channels into separate tracks and you will find near identical waveforms.
Use two identical mics placed one upon another and send the result to an Osmo Pocket and the other to a separate sound recorder and it's chalk and cheese.
For a really portable setup with a small microphone the adapter is useful and help keeps down the extra amount of gear needed but not helpful in any other respect. If you look at the last video I put up and go to the last few minutes where I'm in the middle of a crowd you would find it hard to know who is saying what or where even though the speech is clear enough.
I've built PA systems, PCs, mics, radios and all types of electronic gear over the years and this has me baffled with one other question.
In the video data file, the audio steam is labeled DJI.AAC. Does that mean DJI have their own variation of aac? is it the encoding that can't be read back?
If so there may be hope yet but not from the admins here. It's not their fault and I'm sure some of what we type on this forum gets passed on.
But then again what car ever does the advertised mileage?
Does a washing machine really get rid of all those stains at the described temperatures?
Will Britain ever leave Europe?
Answers on a postcard please.
|
|