Mixing 12-bit and 16-bit audio

btspm · 12th June 2006, 03:32

I'm editing a sequence in Premiere Pro 2.0 that incorporates about 55 minutes of footage with 12-bit audio and 5 minutes of 16-bit footage tacked onto the end. When I export to CCE using debugmode frameserver the video sequence is shortened by about 10 minutes and the 12-bit portion of the video is noticeably and drastically speeded up. The audio exports to .wav just fine and is the correct length.

Is it impossible to mix the two formats in a single sequence? My understanding is that each sequence can have only one audio rate and that premiere will upsample the 12-bit audio. Do I have to recapture and tweak the settings then? Or should I bite the bullet on quality and use Premiere's built in 'export to DVD' encoder?

Any advice would be much appreciated.

gameplaya15143 · 15th June 2006, 19:37

demux the 12bit 32khz audio, and convert it to 16bit 48khz.. then go from there.

I would suggest changing the smaller clip's 16bit audio to 12bit, but I've no clue how to go about that.

smok3 · 15th June 2006, 23:16

well, you could just duplicate the edit and remove the audio completely, then serve that...., or maybe read the frameservers docs even.

btspm · 18th June 2006, 00:21

It looks like I will have to demux since even if I do not export the audio, the video portion of the sequence is still part of a muxed avi even if premiere isn't using the audio.

Thanks for the suggestions.

communist · 18th June 2006, 19:18

Hmm I havent used Premiere 2.0 but in 1.0 / 1.5 you shouldn have to touch different audio clips at all, as they were 'conformed' to one sampling/precision setting at import anyway. You could import anything and they would end up being decoded to a new wav file (if I understand the manual right).
Does 2.0 remove the "audio conforming" step?

12th June 2006, 03:32	#1 \| Link
btspm Registered User Join Date: Jun 2004 Posts: 32	Mixing 12-bit and 16-bit audio I'm editing a sequence in Premiere Pro 2.0 that incorporates about 55 minutes of footage with 12-bit audio and 5 minutes of 16-bit footage tacked onto the end. When I export to CCE using debugmode frameserver the video sequence is shortened by about 10 minutes and the 12-bit portion of the video is noticeably and drastically speeded up. The audio exports to .wav just fine and is the correct length. Is it impossible to mix the two formats in a single sequence? My understanding is that each sequence can have only one audio rate and that premiere will upsample the 12-bit audio. Do I have to recapture and tweak the settings then? Or should I bite the bullet on quality and use Premiere's built in 'export to DVD' encoder? Any advice would be much appreciated.

15th June 2006, 19:37	#2 \| Link
gameplaya15143 Codec Tweaker Join Date: Apr 2006 Location: Earth Posts: 184	demux the 12bit 32khz audio, and convert it to 16bit 48khz.. then go from there. I would suggest changing the smaller clip's 16bit audio to 12bit, but I've no clue how to go about that. __________________ My Site \| SKOSC \| dBpowerAMP \| vfw h264 decoding lame 3.93.1 -q 5 -V 9 -k -Z --nspsytune vorbis -q 0 --advanced-encode-option lowpass_frequency=99

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

15th June 2006, 23:16	#3 \| Link
smok3 videographer Join Date: Oct 2001 Posts: 1,679	well, you could just duplicate the edit and remove the audio completely, then serve that...., or maybe read the frameservers docs even.

18th June 2006, 00:21	#4 \| Link
btspm Registered User Join Date: Jun 2004 Posts: 32	It looks like I will have to demux since even if I do not export the audio, the video portion of the sequence is still part of a muxed avi even if premiere isn't using the audio. Thanks for the suggestions.

18th June 2006, 19:18	#5 \| Link
communist Registered User Join Date: Jul 2003 Posts: 1,141	Hmm I havent used Premiere 2.0 but in 1.0 / 1.5 you shouldn have to touch different audio clips at all, as they were 'conformed' to one sampling/precision setting at import anyway. You could import anything and they would end up being decoded to a new wav file (if I understand the manual right). Does 2.0 remove the "audio conforming" step?