murrayatuptown
22nd September 2013, 22:11
Hi:
I'm a bumbling beginner at video editing. I shoot videos of my daughter's street musician efforts, as well as occasional indoor concerts.
I use inexpensive point&shoot cameras that record the 1280 x 720 variety of HD, saved as .mov files.
Most times, the output need not be anything higher quality that that needed for upload to Facebook or YouTube.
Camera audio quality has been disappointing. I've done two things with inconsistent results. Any suggestions for how to extract for compatibility, then re-sync post-processed audio and video into an MP4 package would be appreciated.
1) Record audio separately with something like a Tascam digital recorder and attempt to sync the audio to the video from which I stripped the audio.
2) Extract the original .mov audio, post process it, if it's at all usable, and try to remux it back in, hoping the identical size allows better sync.
I used YAMB twice with MP4Box, and observed some strange things, like reference to an interleave of 0.500 seconds. One video ignored any attempts to introduce delay between the audio and video, despite the ability to set the delay in milliseconds.
I suspect the components I'm trying to remux are the problem. I'm not even sure what questions to be asking because I don't understand what part isn't working as expected.
Today I realized that extracting audio and saving it as 44.1 k stereo might not be good idea if what the stream info says is 16 kHz 128 kbps uLaw 2:1 mono, so I 'saved as' 16k mono in today's attempt.
There is a third thing I've attempted...hacking the in-camera electret mic in an attempt to implement a Linkwitz mod to prevent overload of the in-camera audio. I've gone to an external jack and a different cheap ECM capsule...some improvement but still trying to analyze new oddities.
Wind noise, Harleys, drums and electric basses are my biggest headaches.
Here's a sample log:
MP3 import - sample rate 16000 - MPEG-2 audio - 1 channel
IsoMedia import - track ID 1 - Video (size 1280 x 720)
Saving to D:\CanidaeSep21_13\100_7222.mp4: 0.500 secs Interleaving
Most recent software tools attempted are MPEGStreamclip to extract audio and trim finished or unedited video, and YAMB/MP4Box to re-mux.
So far, post processing (leveling, EQ, etc.) in Audacity seems to work fine.
Maybe I break something in one of the subsequent steps I perform to the audio.
I've only uploaded one video online that has remuxed audio reinserted into the video. It's a guitar duo that is too fast for most people to notice the sync problem...until the end (!) where there's a huge discrepancy. This is where I noticed the 0.500 second interleave statement regardless of advance or delay attempts. I can post a link to that if it's helpful.
Thank you
Murray
I'm a bumbling beginner at video editing. I shoot videos of my daughter's street musician efforts, as well as occasional indoor concerts.
I use inexpensive point&shoot cameras that record the 1280 x 720 variety of HD, saved as .mov files.
Most times, the output need not be anything higher quality that that needed for upload to Facebook or YouTube.
Camera audio quality has been disappointing. I've done two things with inconsistent results. Any suggestions for how to extract for compatibility, then re-sync post-processed audio and video into an MP4 package would be appreciated.
1) Record audio separately with something like a Tascam digital recorder and attempt to sync the audio to the video from which I stripped the audio.
2) Extract the original .mov audio, post process it, if it's at all usable, and try to remux it back in, hoping the identical size allows better sync.
I used YAMB twice with MP4Box, and observed some strange things, like reference to an interleave of 0.500 seconds. One video ignored any attempts to introduce delay between the audio and video, despite the ability to set the delay in milliseconds.
I suspect the components I'm trying to remux are the problem. I'm not even sure what questions to be asking because I don't understand what part isn't working as expected.
Today I realized that extracting audio and saving it as 44.1 k stereo might not be good idea if what the stream info says is 16 kHz 128 kbps uLaw 2:1 mono, so I 'saved as' 16k mono in today's attempt.
There is a third thing I've attempted...hacking the in-camera electret mic in an attempt to implement a Linkwitz mod to prevent overload of the in-camera audio. I've gone to an external jack and a different cheap ECM capsule...some improvement but still trying to analyze new oddities.
Wind noise, Harleys, drums and electric basses are my biggest headaches.
Here's a sample log:
MP3 import - sample rate 16000 - MPEG-2 audio - 1 channel
IsoMedia import - track ID 1 - Video (size 1280 x 720)
Saving to D:\CanidaeSep21_13\100_7222.mp4: 0.500 secs Interleaving
Most recent software tools attempted are MPEGStreamclip to extract audio and trim finished or unedited video, and YAMB/MP4Box to re-mux.
So far, post processing (leveling, EQ, etc.) in Audacity seems to work fine.
Maybe I break something in one of the subsequent steps I perform to the audio.
I've only uploaded one video online that has remuxed audio reinserted into the video. It's a guitar duo that is too fast for most people to notice the sync problem...until the end (!) where there's a huge discrepancy. This is where I noticed the 0.500 second interleave statement regardless of advance or delay attempts. I can post a link to that if it's helpful.
Thank you
Murray