Clarification needed for handling raw AAC bitstreams without description #832
Comments
ChunMinChang
changed the title
|
Hmm, Chromium doesn't intentionally generate an audio specific config AFAIK. FFmpeg may be doing something under the hood though -- I doubt this works if we end up having to use a platform decoder. I don't know enough about AAC to say if there are forms where AAC can be decoded w/o the ASC. |
Very often you can make a good guess if you have the channel count / sample rate, and get something to decode fine by generating extradata. For example, here's some code that does it: https://searchfox.org/mozilla-central/source/dom/media/platforms/agnostic/bytestreams/Adts.cpp#273-287. Related: #826 (comment) |
moz-wptsync-bot
pushed a commit
to web-platform-tests/wpt
that referenced
this issue
Oct 3, 2024
Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1917721 gecko-commit: c81481c06eac3165b72e7bb2c3040a48b3994a0a gecko-reviewers: media-playback-reviewers, padenot
moz-v2v-gh
pushed a commit
to mozilla/gecko-dev
that referenced
this issue
Oct 4, 2024
…edia-playback-reviewers,padenot Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712
moz-wptsync-bot
pushed a commit
to web-platform-tests/wpt
that referenced
this issue
Oct 4, 2024
Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712 bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1917721 gecko-commit: c81481c06eac3165b72e7bb2c3040a48b3994a0a gecko-reviewers: media-playback-reviewers, padenot
gecko-dev-updater
pushed a commit
to marco-c/gecko-dev-comments-removed
that referenced
this issue
Oct 9, 2024
…edia-playback-reviewers,padenot Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712 UltraBlame original commit: c81481c06eac3165b72e7bb2c3040a48b3994a0a
gecko-dev-updater
pushed a commit
to marco-c/gecko-dev-wordified
that referenced
this issue
Oct 9, 2024
…edia-playback-reviewers,padenot Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712 UltraBlame original commit: c81481c06eac3165b72e7bb2c3040a48b3994a0a
i3roly
pushed a commit
to i3roly/firefox-dynasty
that referenced
this issue
Oct 9, 2024
…edia-playback-reviewers,padenot Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712
ErichDonGubler
pushed a commit
to erichdongubler-mozilla/firefox
that referenced
this issue
Oct 10, 2024
…edia-playback-reviewers,padenot Due to the ongoing discussion in issue 832 [1], which has not yet reached a consensus among all parties, the AAC-without-description WPT introduced in the previous patch should be relocated to the *mozilla* folder under WPT. This ensures that the test remains specific to Mozilla and avoids unintentionally causing failures in other browers' test suites. [1] w3c/webcodecs#832 Differential Revision: https://phabricator.services.mozilla.com/D222712
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
According to the AAC registration in WebCodecs, when the description is not present in AudioDecoderConfig, the bitstream is assumed to be in ADTS format. Otherwise, it's assumed to be in AAC. However, there's a case (https://webvideo-js.pages.dev/) where raw AAC data is provided to the audio decoder without supplying description data in configure beforehand (#826 seems a similar issue for opus).
Currently, Chrome can decode this raw AAC data, and Firefox has patches to support this behavior as well (by generating a AudioSpecificConfig).
This issue can be easily reproduced by adding a test case in audioDecoder-codec-specific.https.any.js WPT that replicates the properties of
MP4_AAC_DATAbut omits the description. I would like to ask if this behavior should be officially mentioned or detailed in the WebCodecs spec. If this use case is something that should be supported, it might be worthwhile to include a corresponding WPT for this scenario.The text was updated successfully, but these errors were encountered: