Mp4Opus - Revision history

MrZeus at 15:36, 18 August 2015

2015-08-18T15:36:49Z

← Older revision		Revision as of 15:36, 18 August 2015
Line 51:		Line 51:

	what about album art? quicktime/mp4/mp3		what about album art? quicktime/mp4/mp3

			[[Category:Opus]]

Tdaede at 23:35, 17 August 2015

2015-08-17T23:35:22Z

← Older revision		Revision as of 23:35, 17 August 2015
Line 1:		Line 1:
	{{draft}}		{{draft}}
	~~This~~ is a draft encapsulation guide for [[Opus]] audio in the mp4 (ISO Base) media container.		[http://vfrmaniac.fushizen.eu/contents/opus_in_isobmff.html Paranoialmaniac's draft] is the current draft encapsulation guide for [[Opus]] audio in the mp4 (ISO Base) media container.

			Below are some of the motivations behind the draft.

	[http://wiki.multimedia.cx/index.php?title=MP4 MP4] already has support for declaring encoder delay and pre-roll.		[http://wiki.multimedia.cx/index.php?title=MP4 MP4] already has support for declaring encoder delay and pre-roll.
Line 8:		Line 10:
	For delay, Daemon404 suggested "whatever l-smash maps encoder delay to."		For delay, Daemon404 suggested "whatever l-smash maps encoder delay to."
	yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list entry.		yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list entry.

	~~See also [http://vfrmaniac.fushizen.eu/contents/opus_in_isobmff.html Paranoialmaniac's draft].~~

	There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.		There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.

Rillian: Link to Paranoialmaniac's draft

2014-09-23T19:09:46Z

Link to Paranoialmaniac's draft

← Older revision		Revision as of 19:09, 23 September 2014
Line 9:		Line 9:
	yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list entry.		yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list entry.

			See also [http://vfrmaniac.fushizen.eu/contents/opus_in_isobmff.html Paranoialmaniac's draft].

	There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.		There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.

Rillian: One edit list entry supports both start and end trim.

2014-09-23T19:07:45Z

One edit list entry supports both start and end trim.

← Older revision		Revision as of 19:07, 23 September 2014
Line 7:		Line 7:

	For delay, Daemon404 suggested "whatever l-smash maps encoder delay to."		For delay, Daemon404 suggested "whatever l-smash maps encoder delay to."
	yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list.		yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list entry.

Derf at 17:12, 21 September 2014

2014-09-21T17:12:23Z

← Older revision		Revision as of 17:12, 21 September 2014
Line 21:		Line 21:

	''AudioRollRecoveryEntry'' - shall have a value of 3840 (80ms * 48k)		''AudioRollRecoveryEntry'' - shall have a value of 3840 (80ms * 48k)

	''AudioSampleEntry'' - hardcoded at 16fp		''AudioSampleEntry'' - hardcoded at 16fp

Derf at 17:11, 21 September 2014

2014-09-21T17:11:54Z

← Older revision		Revision as of 17:11, 21 September 2014
Line 20:		Line 20:
	'''Global Header'''		'''Global Header'''

	''AudioRollRecoveryEntry'' - shall have a value of ~~384~~ (~~8ms~~ * 48k)		''AudioRollRecoveryEntry'' - shall have a value of 3840 (80ms * 48k)
	''AudioSampleEntry'' - hardcoded at 16fp		''AudioSampleEntry'' - hardcoded at 16fp

Koda at 17:00, 21 September 2014

2014-09-21T17:00:13Z

← Older revision		Revision as of 17:00, 21 September 2014
Line 14:		Line 14:
	Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?		Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?

	Possible registration process, send an email to http://mp4ra.org/request.html Possible registrar is '~~opus~~'.		Possible registration process, send an email to http://mp4ra.org/request.html. Possible registrar is 'Opus'.

	Internally everything is resampled at 48000, this is always output by the decoder, floating point numbers. But the original sample rate is stored so that the decoder can act upon it. Atom AudioSampleEntry will have 48 and the original one will be stored in the codec's one.		Internally everything is resampled at 48000, this is always output by the decoder, floating point numbers. But the original sample rate is stored so that the decoder can act upon it. Atom AudioSampleEntry will have 48 and the original one will be stored in the codec's one.

	~~AudioRollRecoveryEntry - shall have a value of 384 (8ms * 48k)~~		'''Global Header'''
	~~AudioSampleEntry - hardcoded at 16fp~~

	~~descriptor~~ - ~~same as ogg rather than as ts, to keep things simple~~		''AudioRollRecoveryEntry'' - shall have a value of 384 (8ms * 48k)
			''AudioSampleEntry'' - hardcoded at 16fp

	~~channel count~~ - ~~already included~~		''descriptor'' - same as ogg rather than as ts, to keep things simple
	~~pre skip - already included~~

	Gain - volume atom? unused in practice - oggheader - not in ts (TODO?)		''channel count'' - already included
			''pre skip'' - already included

			''Gain'' - volume atom? unused in practice - oggheader - not in ts (TODO?)
	when you decode samples you're supposed to multiply against this value, so that decoder can apply post volume		when you decode samples you're supposed to multiply against this value, so that decoder can apply post volume
	Reusing the one in ogg.		Reusing the one in ogg.

	mapping family (with vorbis mapping)		''mapping family'' (with vorbis mapping)
	0 - mono/stereo no channel config		- mono/stereo no channel config
	1 - specify # channel		- specify # channel
	2 - map to the # ouput		- map to the # ouput

	audio channel layout https://developer.apple.com/library/mac/documentation/musicaudio/reference/CoreAudioDataTypesRef/Reference/reference.html - too complex		audio channel layout https://developer.apple.com/library/mac/documentation/musicaudio/reference/CoreAudioDataTypesRef/Reference/reference.html - too complex
	plug it from ogg and put it in our custom atom		plug it from ogg and put it in our custom atom

	Things to put in custom atom		'''Things to put in custom atom'''
	- input sr		- input sr
	- output gain		- output gain
	- channel mappaing		- channel mappaing
			- channel count (for backup)

			'''Opus''' is the name of the atom, like in TS

			what about album art? quicktime/mp4/mp3

Koda at 16:56, 21 September 2014

2014-09-21T16:56:05Z

← Older revision		Revision as of 16:56, 21 September 2014
Line 13:		Line 13:

	Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?		Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?

			Possible registration process, send an email to http://mp4ra.org/request.html Possible registrar is 'opus'.

			Internally everything is resampled at 48000, this is always output by the decoder, floating point numbers. But the original sample rate is stored so that the decoder can act upon it. Atom AudioSampleEntry will have 48 and the original one will be stored in the codec's one.

			AudioRollRecoveryEntry - shall have a value of 384 (8ms * 48k)
			AudioSampleEntry - hardcoded at 16fp

			descriptor - same as ogg rather than as ts, to keep things simple

			channel count - already included
			pre skip - already included

			Gain - volume atom? unused in practice - oggheader - not in ts (TODO?)
			when you decode samples you're supposed to multiply against this value, so that decoder can apply post volume
			Reusing the one in ogg.

			mapping family (with vorbis mapping)
			0 - mono/stereo no channel config
			1 - specify # channel
			2 - map to the # ouput
			audio channel layout https://developer.apple.com/library/mac/documentation/musicaudio/reference/CoreAudioDataTypesRef/Reference/reference.html - too complex
			plug it from ogg and put it in our custom atom

			Things to put in custom atom
			- input sr
			- output gain
			- channel mappaing

Tmatth: added link to multimedia wiki entry on mp4

2014-07-09T17:58:28Z

added link to multimedia wiki entry on mp4

← Older revision		Revision as of 17:58, 9 July 2014
Line 2:		Line 2:
	This is a draft encapsulation guide for [[Opus]] audio in the mp4 (ISO Base) media container.		This is a draft encapsulation guide for [[Opus]] audio in the mp4 (ISO Base) media container.

	MP4 already has support for declaring encoder delay and pre-roll.		[http://wiki.multimedia.cx/index.php?title=MP4 MP4] already has support for declaring encoder delay and pre-roll.

	For pre-roll I believe we can use 'AudioRollRecoveryEntry' for pre-roll.		For pre-roll I believe we can use 'AudioRollRecoveryEntry' for pre-roll.

Rillian: Mention edit lists for pre-skip

2014-07-09T17:45:33Z

Mention edit lists for pre-skip

← Older revision		Revision as of 17:45, 9 July 2014
Line 5:		Line 5:

	For pre-roll I believe we can use 'AudioRollRecoveryEntry' for pre-roll.		For pre-roll I believe we can use 'AudioRollRecoveryEntry' for pre-roll.

			For delay, Daemon404 suggested "whatever l-smash maps encoder delay to."
			yusuke says, "ISO and Apple recommend the use of edit list for removing priming samples from the presentation." libavformat's demuxer supports one edit list.


	There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.		There's some work on codec-independent channel mapping, downmix and dynamic range control as part of [http://mpeg.chiariglione.org/standards/mpeg-4/iso-base-media-file-format/text-isoiec-14496-122012-pdam-4-enhanced-audio-support ISO 14496-12 Amd4] We might be able use some of that, but it doesn't support the Opus case of needing to indicated which streams are coupled pairs. We'll still need to define our own extension for this.

	Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?		Question: Better to reuse the channel mapping header entirely, or just report the coupled stream count and use the downmix table to do the mapping?