eden-emu/opus
17
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

First set of changes for Gen-art review

Browse source
This commit is contained in:
Jean-Marc Valin 2012-05-11 13:41:41 -04:00
parent 516c980585
commit e8c437c432
1 changed files with 42 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

50
doc/draft-ietf-codec-opus.xml
View file

@ -83,7 +83,7 @@ It is composed of a linear
prediction (LP)-based <xref target="LPC"/> layer and a Modified Discrete Cosine Transform prediction (LP)-based <xref target="LPC"/> layer and a Modified Discrete Cosine Transform
(MDCT)-based <xref target="MDCT"/> layer. (MDCT)-based <xref target="MDCT"/> layer.
The main idea behind using two layers is that in speech, linear prediction The main idea behind using two layers is that in speech, linear prediction
techniques (such as CELP) code low frequencies more efficiently than transform techniques (such as Code-Excited Linear Prediction, or CELP) code low frequencies more efficiently than transform
(e.g., MDCT) domain techniques, while the situation is reversed for music and (e.g., MDCT) domain techniques, while the situation is reversed for music and
higher speech frequencies. higher speech frequencies.
Thus a codec with both layers available can operate over a wider range than Thus a codec with both layers available can operate over a wider range than
@ -150,7 +150,8 @@ E.g., the text will explicitly indicate any shifts required after a
<t> <t>
Expressions, where included in the text, follow C operator rules and Expressions, where included in the text, follow C operator rules and
precedence, with the exception that the syntax "x**y" indicates x raised to precedence, with the exception that the syntax "x**y" indicates x raised to
the power y. the power y. Throughout this document, the term "byte" is defined to include 8 bits,
i.e. an octet.
The text also makes use of the following functions: The text also makes use of the following functions:
</t> </t>
@ -221,6 +222,12 @@ Examples:
</t> </t>
</section> </section>
<section anchor="floor" toc="exclude" title="floor(x)">
<t>
Largest integer z such that z &lt;= x.
</t>
</section>
</section> </section>
</section> </section>
@ -279,7 +286,7 @@ It supports NB, MB, or WB audio and frame sizes from 10&nbsp;ms to 60&nbsp;ms,
and requires an additional 5&nbsp;ms look-ahead for noise shaping estimation. and requires an additional 5&nbsp;ms look-ahead for noise shaping estimation.
A small additional delay (up to 1.5 ms) may be required for sampling rate A small additional delay (up to 1.5 ms) may be required for sampling rate
conversion. conversion.
Like Vorbis and many other modern codecs, SILK is inherently designed for Like Vorbis <xref target='Vorbis-website'/> and many other modern codecs, SILK is inherently designed for
variable-bitrate (VBR) coding, though the encoder can also produce variable-bitrate (VBR) coding, though the encoder can also produce
constant-bitrate (CBR) streams. constant-bitrate (CBR) streams.
The version of SILK used in Opus is substantially modified from, and not The version of SILK used in Opus is substantially modified from, and not
@ -477,7 +484,8 @@ is required. There are two main reasons to operate in CBR mode:
When low-latency transmission is required over a relatively slow connection, then When low-latency transmission is required over a relatively slow connection, then
constrained VBR can also be used. This uses VBR in a way that simulates a constrained VBR can also be used. This uses VBR in a way that simulates a
"bit reservoir" and is equivalent to what MP3 and AAC call CBR (i.e. not true "bit reservoir" and is equivalent to what MP3 (MPEG 1, Layer 3) and
AAC (Advanced Audio Coding) call CBR (i.e. not true
CBR due to the bit reservoir). CBR due to the bit reservoir).
</t> </t>
</section> </section>
@ -507,7 +515,8 @@ A single packet may contain multiple audio frames, so long as they share a
This section describes the possible combinations of these parameters and the This section describes the possible combinations of these parameters and the
internal framing used to pack multiple frames into a single packet. internal framing used to pack multiple frames into a single packet.
This framing is not self-delimiting. This framing is not self-delimiting.
Instead, it assumes that a higher layer (such as UDP or RTP or Ogg or Matroska) Instead, it assumes that a higher layer (such as UDP or RTP <xref target='RFC3550'/>
or Ogg <xref target='RFC3533'/> or Matroska <xref target='Matroska-website'/>)
will communicate the length, in bytes, of the packet, and it uses this will communicate the length, in bytes, of the packet, and it uses this
information to reduce the framing overhead in the packet itself. information to reduce the framing overhead in the packet itself.
A decoder implementation MUST support the framing described in this section. A decoder implementation MUST support the framing described in this section.
@ -1000,7 +1009,8 @@ stream | Range |---+ +---------+ +------------+ /---\ Audio
<section anchor="range-decoder" title="Range Decoder"> <section anchor="range-decoder" title="Range Decoder">
<t> <t>
Opus uses an entropy coder based on <xref target="range-coding"></xref>, Opus uses an entropy coder based on range coding <xref target="range-coding"></xref>
<xref target="Nigel79"></xref>,
which is itself a rediscovery of the FIFO arithmetic code introduced by <xref target="coding-thesis"></xref>. which is itself a rediscovery of the FIFO arithmetic code introduced by <xref target="coding-thesis"></xref>.
It is very similar to arithmetic encoding, except that encoding is done with It is very similar to arithmetic encoding, except that encoding is done with
digits in any base instead of with bits, digits in any base instead of with bits,
@ -6148,7 +6158,7 @@ The procedure in <xref target="encoder-finalizing"/> does this in a way that
The function ec_enc_uint() (entenc.c) encodes one of ft equiprobable symbols in The function ec_enc_uint() (entenc.c) encodes one of ft equiprobable symbols in
the range 0 to (ft&nbsp;-&nbsp;1), inclusive, each with a frequency of 1, the range 0 to (ft&nbsp;-&nbsp;1), inclusive, each with a frequency of 1,
where ft may be as large as (2**32&nbsp;-&nbsp;1). where ft may be as large as (2**32&nbsp;-&nbsp;1).
Like the decoder (see <xref target="ec_dec_uint"/>), it splits it splits up the Like the decoder (see <xref target="ec_dec_uint"/>), it splits up the
value into a range coded symbol representing up to 8 of the high bits, and, if value into a range coded symbol representing up to 8 of the high bits, and, if
necessary, raw bits representing the remainder of the value. necessary, raw bits representing the remainder of the value.
</t> </t>
@ -7489,6 +7499,9 @@ name of work, or endorsement information.</t>
<format type='TXT' target='http://tools.ietf.org/rfc/rfc6366.txt' /> <format type='TXT' target='http://tools.ietf.org/rfc/rfc6366.txt' />
</reference> </reference>
<?rfc include="http://xml.resource.org/public/rfc/bibxml/reference.RFC.3550.xml"?>
<?rfc include="http://xml.resource.org/public/rfc/bibxml/reference.RFC.3533.xml"?>
<reference anchor='SILK' target='http://developer.skype.com/silk'> <reference anchor='SILK' target='http://developer.skype.com/silk'>
<front> <front>
<title>SILK Speech Codec</title> <title>SILK Speech Codec</title>
@ -7590,7 +7603,7 @@ Robust and Efficient Quantization of Speech LSP Parameters Using Structured Vect
<format type='TXT' octets='110393' target='ftp://ftp.isi.edu/in-notes/rfc3552.txt' /> <format type='TXT' octets='110393' target='ftp://ftp.isi.edu/in-notes/rfc3552.txt' />
</reference> </reference>
<reference anchor="range-coding"> <reference anchor="Nigel79">
<front> <front>
<title>Range encoding: An algorithm for removing redundancy from a digitised message</title> <title>Range encoding: An algorithm for removing redundancy from a digitised message</title>
<author initials="G." surname="Nigel" fullname=""><organization/></author> <author initials="G." surname="Nigel" fullname=""><organization/></author>
@ -7654,6 +7667,20 @@ Robust and Efficient Quantization of Speech LSP Parameters Using Structured Vect
</front> </front>
</reference> </reference>
<reference anchor="Vorbis-website" target="http://vorbis.com/">
<front>
<title>Vorbis website</title>
<author></author>
</front>
</reference>
<reference anchor="Matroska-website" target="http://matroska.org/">
<front>
<title>Matroska website</title>
<author></author>
</front>
</reference>
<reference anchor="Vectors-website" target="http://opus-codec.org/testvectors/"> <reference anchor="Vectors-website" target="http://opus-codec.org/testvectors/">
<front> <front>
<title>Opus Testvectors (webside)</title> <title>Opus Testvectors (webside)</title>
@ -7668,6 +7695,13 @@ Robust and Efficient Quantization of Speech LSP Parameters Using Structured Vect
</front> </front>
</reference> </reference>
<reference anchor="range-coding" target="http://en.wikipedia.org/wiki/Range_coding">
<front>
<title>Range Coding</title>
<author><organization>Wikipedia</organization></author>
</front>
</reference>
<reference anchor="Hadamard" target="http://en.wikipedia.org/wiki/Hadamard_transform"> <reference anchor="Hadamard" target="http://en.wikipedia.org/wiki/Hadamard_transform">
<front> <front>
<title>Hadamard Transform</title> <title>Hadamard Transform</title>