[Mp4-tech] Re: Can I use PCM or ADPCM as audio stream of .mp4 fil e?

[Ralph Sperschneider]

John Lam wrote:
> Hi,
> 
> Thank you for your help!
> Is it sure? Did you try it? As I don't have 14496-3 spec, I cannot verify it myself and so I don't know how to pack 16-bit PCM in audio stream.
> 
> 
> Alan Thomson wrote:
> 
> 
>>Hi -
>>
>>14496-3:2001 subpart 1 paras 1.5, 1.6 defines the Audio Object Type "Null"
>>with Type ID 0 as 16-bit PCM, if that helps?
>>
>>rgds,
>>Alan.
>>
>>>-----Original Message-----
>>>From: Dave Singer [mailto:singer apple.com]
>>>Sent: 13 January 2024 21:57
>>>To: John Lam; mp4-tech
>>>Subject: Re: [Mp4-tech] Can I use PCM or ADPCM as audio stream of .mp4
>>>file?
>>>
>>>
>>>At 12:16 PM +0800 1/9/04, John Lam wrote:
>>>
>>>>Hi all,
>>>>
>>>>I'm now working on construct .mp4 file with both visual and audio
>>>>stream. Can audio stream data be non-compressed PCM (16 bits) or IMA
>>>>ADPCM (4 bits) data? If so, how to fill the whole 'trak' atom and its
>>>>sub-atom such as 'hdlr'? If not, what file format should I use for
>>>>packing MPEG4 visual stream simple profile confirmed to
>>>
>>>ISO/IEC 14496-2
>>>
>>>>and audio stream is either non-compressed PCM or IMA ADPCM together?
>>>>Thank you for your help!
>>>
>>>MPEG-4 doesn't define an uncompressed storage, except that there is
>>>(I think) some way of labelling a stream as AAC but storing it
>>>uncompressed.  However, the Motion JPEG 2000 spec (part 3) does
>>>define uncompressed 8 and 16 bit storage;  and MJ2 uses the same file
>>>format as MP4.
>>>
>>>Hope this helps.
>>>
>>>
>>>>--
>>>>Best Regards,
>>>>
>>>>John Lam
>>>>
>>>>
>>>>_______________________________________________
>>>>Mp4-tech mailing list
>>>>Mp4-tech lists.mpegif.org
>>>>http://lists.mpegif.org/mailman/listinfo/mp4-tech
>>>>
>>>>Note: Conduct on the mailing list is subject to the Antitrust
>>>>guidelines found at
>>>>http://www.mpegif.org/public/documents/vault/mp-out-30042-Ant
>>
>>itrust.php
>>
>>--
>>David Singer
>>Apple Computer/QuickTime
>>_______________________________________________
>>Mp4-tech mailing list
>>Mp4-tech lists.mpegif.org
>>http://lists.mpegif.org/mailman/listinfo/mp4-tech
>>
>>Note: Conduct on the mailing list is subject to the Antitrust guidelines
>>found at
>>http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>>_______________________________________________
>>Mp4-tech mailing list
>>Mp4-tech lists.mpegif.org
>>http://lists.mpegif.org/mailman/listinfo/mp4-tech
>>
>>Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> 
> 
> --
> Best Regards,
> 
> ___________________________________________________________________
> John Lam <johnlam fujitsu-hkdc.com>
> Phone : (852)2736-3232 Direct : (852)2737-9655 Fax : (852)2314-4207
> 
> Fujitsu Microelectronics Pacific Asia Ltd., Design Centre
> 
> 
> _______________________________________________
> Mp4-tech mailing list
> Mp4-tech lists.mpegif.org
> http://lists.mpegif.org/mailman/listinfo/mp4-tech
> 
> Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php

Dear John, Alan, Dave,
I found the following in the standards:
Part 3: 1.5.1.2.1	NULL Object
The NULL object provides the possibility to feed raw PCM data directly to the
audio compositor. No decoding is involved, However, an audio object descriptor
is used to specify the sampling rate and the audio channel configuration.
Part 3: 1.6.2.2.1	Overview
For the NULL object the payload shall be 16 bit signed integer in the range from
-32768 to +32767.
Part 4: 5.6.2	Null
The NULL object type provides the possibility to feed raw PCM data directly into
the audio compositor. No decoding is involved. The sampling rate and the audio
channel configuration is specified by the AudioSpecificConfig.
Note: The AOT 0 has been specified for testing purposes only. Therefore it does 
not appear in any profile, and its specification is not very sophisticated. It 
seems that no information is given about the number of samples per access
unit. I wonder whether this would be of relevance.
Best regards,
Ralph
-- 
Dipl.-Ing. Ralph Sperschneider  | Phone: +49 9131 776 344
FhG IIS                         | Fax:   +49 9131 776 398
Am Wolfsmantel 33               | mailto:ralph.sperschneider iis.fraunhofer.de
D 91058 Erlangen                | http://www.iis.fhg.de/amm/