• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1<?xml version="1.0" encoding="UTF-8"?>
2<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
3<!ENTITY rfc2119 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.2119.xml'>
4<!ENTITY rfc3389 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3389.xml'>
5<!ENTITY rfc3550 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3550.xml'>
6<!ENTITY rfc3711 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3711.xml'>
7<!ENTITY rfc3551 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3551.xml'>
8<!ENTITY rfc6838 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.6838.xml'>
9<!ENTITY rfc4855 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.4855.xml'>
10<!ENTITY rfc4566 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.4566.xml'>
11<!ENTITY rfc4585 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.4585.xml'>
12<!ENTITY rfc3264 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3264.xml'>
13<!ENTITY rfc2974 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.2974.xml'>
14<!ENTITY rfc2326 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.2326.xml'>
15<!ENTITY rfc3555 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.3555.xml'>
16<!ENTITY rfc5124 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.5124.xml'>
17<!ENTITY rfc5405 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.5405.xml'>
18<!ENTITY rfc5576 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.5576.xml'>
19<!ENTITY rfc6562 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.6562.xml'>
20<!ENTITY rfc6716 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.6716.xml'>
21<!ENTITY rfc7202 PUBLIC '' 'http://xml.resource.org/public/rfc/bibxml/reference.RFC.7202.xml'>
22<!ENTITY nbsp "&#160;">
23  ]>
24
25  <rfc category="std" ipr="trust200902" docName="draft-ietf-payload-rtp-opus-11">
26<?xml-stylesheet type='text/xsl' href='rfc2629.xslt' ?>
27
28<?rfc strict="yes" ?>
29<?rfc toc="yes" ?>
30<?rfc tocdepth="3" ?>
31<?rfc tocappendix='no' ?>
32<?rfc tocindent='yes' ?>
33<?rfc symrefs="yes" ?>
34<?rfc sortrefs="yes" ?>
35<?rfc compact="no" ?>
36<?rfc subcompact="yes" ?>
37<?rfc iprnotified="yes" ?>
38
39  <front>
40    <title abbrev="RTP Payload Format for Opus">
41      RTP Payload Format for the Opus Speech and Audio Codec
42    </title>
43
44    <author fullname="Julian Spittka" initials="J." surname="Spittka">
45      <address>
46        <email>jspittka@gmail.com</email>
47      </address>
48    </author>
49
50    <author initials='K.' surname='Vos' fullname='Koen Vos'>
51      <organization>vocTone</organization>
52      <address>
53        <postal>
54          <street></street>
55          <code></code>
56          <city></city>
57          <region></region>
58          <country></country>
59        </postal>
60        <email>koenvos74@gmail.com</email>
61      </address>
62    </author>
63
64    <author initials="JM" surname="Valin" fullname="Jean-Marc Valin">
65      <organization>Mozilla</organization>
66      <address>
67        <postal>
68          <street>331 E. Evelyn Avenue</street>
69          <city>Mountain View</city>
70          <region>CA</region>
71          <code>94041</code>
72          <country>USA</country>
73        </postal>
74        <email>jmvalin@jmvalin.ca</email>
75      </address>
76    </author>
77
78    <date day='14' month='April' year='2015' />
79
80    <abstract>
81      <t>
82        This document defines the Real-time Transport Protocol (RTP) payload
83        format for packetization of Opus encoded
84        speech and audio data necessary to integrate the codec in the
85        most compatible way. It also provides an applicability statement
86        for the use of Opus over RTP. Further, it describes media type registrations
87        for the RTP payload format.
88      </t>
89    </abstract>
90  </front>
91
92  <middle>
93    <section title='Introduction'>
94      <t>
95        Opus <xref target="RFC6716"/> is a speech and audio codec developed within the
96        IETF Internet Wideband Audio Codec working group. The codec
97        has a very low algorithmic delay and it
98        is highly scalable in terms of audio bandwidth, bitrate, and
99        complexity. Further, it provides different modes to efficiently encode speech signals
100        as well as music signals, thus making it the codec of choice for
101        various applications using the Internet or similar networks.
102      </t>
103      <t>
104        This document defines the Real-time Transport Protocol (RTP)
105        <xref target="RFC3550"/> payload format for packetization
106        of Opus encoded speech and audio data necessary to
107        integrate Opus in the
108        most compatible way. It also provides an applicability statement
109        for the use of Opus over RTP.
110        Further, it describes media type registrations for
111        the RTP payload format.
112      </t>
113    </section>
114
115    <section title='Conventions, Definitions and Acronyms used in this document'>
116      <t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
117      "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
118      document are to be interpreted as described in <xref target="RFC2119"/>.</t>
119      <t>
120      <list style='hanging'>
121          <t hangText="audio bandwidth:"> The range of audio frequecies being coded</t>
122          <t hangText="CBR:"> Constant bitrate</t>
123          <t hangText="CPU:"> Central Processing Unit</t>
124          <t hangText="DTX:"> Discontinuous transmission</t>
125          <t hangText="FEC:"> Forward error correction</t>
126          <t hangText="IP:"> Internet Protocol</t>
127          <t hangText="samples:"> Speech or audio samples (per channel)</t>
128          <t hangText="SDP:"> Session Description Protocol</t>
129          <t hangText="VBR:"> Variable bitrate</t>
130      </list>
131      </t>
132        <t>
133          Throughout this document, we refer to the following definitions:
134        </t>
135          <texttable anchor='bandwidth_definitions'>
136            <ttcol align='center'>Abbreviation</ttcol>
137            <ttcol align='center'>Name</ttcol>
138            <ttcol align='center'>Audio Bandwidth (Hz)</ttcol>
139            <ttcol align='center'>Sampling Rate (Hz)</ttcol>
140            <c>NB</c>
141            <c>Narrowband</c>
142            <c>0 - 4000</c>
143            <c>8000</c>
144
145            <c>MB</c>
146            <c>Mediumband</c>
147            <c>0 - 6000</c>
148            <c>12000</c>
149
150            <c>WB</c>
151            <c>Wideband</c>
152            <c>0 - 8000</c>
153            <c>16000</c>
154
155            <c>SWB</c>
156            <c>Super-wideband</c>
157            <c>0 - 12000</c>
158            <c>24000</c>
159
160            <c>FB</c>
161            <c>Fullband</c>
162            <c>0 - 20000</c>
163            <c>48000</c>
164
165            <postamble>
166              Audio bandwidth naming
167            </postamble>
168          </texttable>
169    </section>
170
171    <section title='Opus Codec'>
172      <t>
173        Opus encodes speech
174        signals as well as general audio signals. Two different modes can be
175        chosen, a voice mode or an audio mode, to allow the most efficient coding
176        depending on the type of the input signal, the sampling frequency of the
177        input signal, and the intended application.
178      </t>
179
180      <t>
181        The voice mode allows efficient encoding of voice signals at lower bit
182        rates while the audio mode is optimized for general audio signals at medium and
183        higher bitrates.
184      </t>
185
186      <t>
187        Opus is highly scalable in terms of audio
188        bandwidth, bitrate, and complexity. Further, Opus allows
189        transmitting stereo signals with in-band signaling in the bit-stream.
190      </t>
191
192      <section title='Network Bandwidth'>
193          <t>
194            Opus supports bitrates from 6&nbsp;kb/s to 510&nbsp;kb/s.
195            The bitrate can be changed dynamically within that range.
196            All
197            other parameters being
198            equal, higher bitrates result in higher audio quality.
199          </t>
200          <section title='Recommended Bitrate' anchor='bitrate_by_bandwidth'>
201          <t>
202            For a frame size of
203            20&nbsp;ms, these
204            are the bitrate "sweet spots" for Opus in various configurations:
205
206          <list style="symbols">
207            <t>8-12 kb/s for NB speech,</t>
208            <t>16-20 kb/s for WB speech,</t>
209            <t>28-40 kb/s for FB speech,</t>
210            <t>48-64 kb/s for FB mono music, and</t>
211            <t>64-128 kb/s for FB stereo music.</t>
212          </list>
213        </t>
214      </section>
215        <section title='Variable versus Constant Bitrate'  anchor='variable-vs-constant-bitrate'>
216          <t>
217            For the same average bitrate, variable bitrate (VBR) can achieve higher audio quality
218            than constant bitrate (CBR). For the majority of voice transmission applications, VBR
219            is the best choice. One reason for choosing CBR is the potential
220            information leak that <spanx style='emph'>might</spanx> occur when encrypting the
221            compressed stream. See <xref target="RFC6562"/> for guidelines on when VBR is
222            appropriate for encrypted audio communications. In the case where an existing
223            VBR stream needs to be converted to CBR for security reasons, then the Opus padding
224            mechanism described in <xref target="RFC6716"/> is the RECOMMENDED way to achieve padding
225            because the RTP padding bit is unencrypted.</t>
226
227          <t>
228            The bitrate can be adjusted at any point in time. To avoid congestion,
229            the average bitrate SHOULD NOT exceed the available
230            network bandwidth. If no target bitrate is specified, the bitrates specified in
231            <xref target='bitrate_by_bandwidth'/> are RECOMMENDED.
232          </t>
233
234        </section>
235
236        <section title='Discontinuous Transmission (DTX)'>
237
238          <t>
239            Opus can, as described in <xref target='variable-vs-constant-bitrate'/>,
240            be operated with a variable bitrate. In that case, the encoder will
241            automatically reduce the bitrate for certain input signals, like periods
242            of silence. When using continuous transmission, it will reduce the
243            bitrate when the characteristics of the input signal permit, but
244            will never interrupt the transmission to the receiver. Therefore, the
245            received signal will maintain the same high level of audio quality over the
246            full duration of a transmission while minimizing the average bit
247            rate over time.
248          </t>
249
250          <t>
251            In cases where the bitrate of Opus needs to be reduced even
252            further or in cases where only constant bitrate is available,
253            the Opus encoder can use discontinuous
254            transmission (DTX), where parts of the encoded signal that
255            correspond to periods of silence in the input speech or audio signal
256            are not transmitted to the receiver. A receiver can distinguish
257            between DTX and packet loss by looking for gaps in the sequence
258            number, as described by Section 4.1
259            of&nbsp;<xref target="RFC3551"/>.
260          </t>
261
262          <t>
263            On the receiving side, the non-transmitted parts will be handled by a
264            frame loss concealment unit in the Opus decoder which generates a
265            comfort noise signal to replace the non transmitted parts of the
266            speech or audio signal. Use of <xref target="RFC3389"/> Comfort
267            Noise (CN) with Opus is discouraged.
268            The transmitter MUST drop whole frames only,
269            based on the size of the last transmitted frame,
270            to ensure successive RTP timestamps differ by a multiple of 120 and
271            to allow the receiver to use whole frames for concealment.
272          </t>
273
274          <t>
275            DTX can be used with both variable and constant bitrate.
276            It will have a slightly lower speech or audio
277            quality than continuous transmission. Therefore, using continuous
278            transmission is RECOMMENDED unless constraints on available network bandwidth
279            are severe.
280          </t>
281
282        </section>
283
284        </section>
285
286      <section title='Complexity'>
287
288        <t>
289          Complexity of the encoder can be scaled to optimize for CPU resources in real-time, mostly as
290          a trade-off between audio quality and bitrate. Also, different modes of Opus have different complexity.
291        </t>
292
293      </section>
294
295      <section title="Forward Error Correction (FEC)">
296
297        <t>
298          The voice mode of Opus allows for embedding "in-band" forward error correction (FEC)
299          data into the Opus bit stream. This FEC scheme adds
300          redundant information about the previous packet (N-1) to the current
301          output packet N. For
302          each frame, the encoder decides whether to use FEC based on (1) an
303          externally-provided estimate of the channel's packet loss rate; (2) an
304          externally-provided estimate of the channel's capacity; (3) the
305          sensitivity of the audio or speech signal to packet loss; (4) whether
306          the receiving decoder has indicated it can take advantage of "in-band"
307          FEC information. The decision to send "in-band" FEC information is
308          entirely controlled by the encoder and therefore no special precautions
309          for the payload have to be taken.
310        </t>
311
312        <t>
313          On the receiving side, the decoder can take advantage of this
314          additional information when it loses a packet and the next packet
315          is available.  In order to use the FEC data, the jitter buffer needs
316          to provide access to payloads with the FEC data.
317          Instead of performing loss concealment for a missing packet, the
318          receiver can then configure its decoder to decode the FEC data from the next packet.
319        </t>
320
321        <t>
322          Any compliant Opus decoder is capable of ignoring
323          FEC information when it is not needed, so encoding with FEC cannot cause
324          interoperability problems.
325          However, if FEC cannot be used on the receiving side, then FEC
326          SHOULD NOT be used, as it leads to an inefficient usage of network
327          resources. Decoder support for FEC SHOULD be indicated at the time a
328          session is set up.
329        </t>
330
331      </section>
332
333      <section title='Stereo Operation'>
334
335        <t>
336          Opus allows for transmission of stereo audio signals. This operation
337          is signaled in-band in the Opus bit-stream and no special arrangement
338          is needed in the payload format. An
339          Opus decoder is capable of handling a stereo encoding, but an
340          application might only be capable of consuming a single audio
341          channel.
342        </t>
343        <t>
344          If a decoder cannot take advantage of the benefits of a stereo signal
345          this SHOULD be indicated at the time a session is set up. In that case
346          the sending side SHOULD NOT send stereo signals as it leads to an
347          inefficient usage of network resources.
348        </t>
349
350      </section>
351
352    </section>
353
354    <section title='Opus RTP Payload Format' anchor='opus-rtp-payload-format'>
355      <t>The payload format for Opus consists of the RTP header and Opus payload
356      data.</t>
357      <section title='RTP Header Usage'>
358        <t>The format of the RTP header is specified in <xref target="RFC3550"/>.
359        The use of the fields of the RTP header by the Opus payload format is
360        consistent with that specification.</t>
361
362        <t>The payload length of Opus is an integer number of octets and
363        therefore no padding is necessary. The payload MAY be padded by an
364        integer number of octets according to <xref target="RFC3550"/>,
365        although the Opus internal padding is preferred.</t>
366
367        <t>The timestamp, sequence number, and marker bit (M) of the RTP header
368        are used in accordance with Section 4.1
369        of&nbsp;<xref target="RFC3551"/>.</t>
370
371        <t>The RTP payload type for Opus is to be assigned dynamically.</t>
372
373        <t>The receiving side MUST be prepared to receive duplicate RTP
374        packets. The receiver MUST provide at most one of those payloads to the
375        Opus decoder for decoding, and MUST discard the others.</t>
376
377        <t>Opus supports 5 different audio bandwidths, which can be adjusted during
378        a stream.
379        The RTP timestamp is incremented with a 48000 Hz clock rate
380        for all modes of Opus and all sampling rates.
381        The unit
382        for the timestamp is samples per single (mono) channel. The RTP timestamp corresponds to the
383        sample time of the first encoded sample in the encoded frame.
384        For data encoded with sampling rates other than 48000 Hz,
385	the sampling rate has to be adjusted to 48000 Hz.</t>
386
387      </section>
388
389      <section title='Payload Structure'>
390        <t>
391          The Opus encoder can output encoded frames representing 2.5, 5, 10, 20,
392          40, or 60&nbsp;ms of speech or audio data. Further, an arbitrary number of frames can be
393          combined into a packet, up to a maximum packet duration representing
394          120&nbsp;ms of speech or audio data. The grouping of one or more Opus
395          frames into a single Opus packet is defined in Section&nbsp;3 of
396          <xref target="RFC6716"/>. An RTP payload MUST contain exactly one
397          Opus packet as defined by that document.
398        </t>
399
400        <t><xref target='payload-structure'/> shows the structure combined with the RTP header.</t>
401
402        <figure anchor="payload-structure"
403                title="Packet structure with RTP header">
404          <artwork align="center">
405            <![CDATA[
406+----------+--------------+
407|RTP Header| Opus Payload |
408+----------+--------------+
409           ]]>
410          </artwork>
411        </figure>
412
413        <t>
414          <xref target='opus-packetization'/> shows supported frame sizes in
415          milliseconds of encoded speech or audio data for the speech and audio modes
416          (Mode) and sampling rates (fs) of Opus and shows how the timestamp is
417          incremented for packetization (ts incr). If the Opus encoder
418          outputs multiple encoded frames into a single packet, the timestamp
419          increment is the sum of the increments for the individual frames.
420        </t>
421
422        <texttable anchor='opus-packetization' title="Supported Opus frame
423         sizes and timestamp increments marked with an o. Unsupported marked with an x.">
424            <ttcol align='center'>Mode</ttcol>
425            <ttcol align='center'>fs</ttcol>
426            <ttcol align='center'>2.5</ttcol>
427            <ttcol align='center'>5</ttcol>
428            <ttcol align='center'>10</ttcol>
429            <ttcol align='center'>20</ttcol>
430            <ttcol align='center'>40</ttcol>
431            <ttcol align='center'>60</ttcol>
432            <c>ts incr</c>
433            <c>all</c>
434            <c>120</c>
435            <c>240</c>
436            <c>480</c>
437            <c>960</c>
438            <c>1920</c>
439            <c>2880</c>
440            <c>voice</c>
441            <c>NB/MB/WB/SWB/FB</c>
442            <c>x</c>
443            <c>x</c>
444            <c>o</c>
445            <c>o</c>
446            <c>o</c>
447            <c>o</c>
448            <c>audio</c>
449            <c>NB/WB/SWB/FB</c>
450            <c>o</c>
451            <c>o</c>
452            <c>o</c>
453            <c>o</c>
454            <c>x</c>
455            <c>x</c>
456          </texttable>
457
458      </section>
459
460    </section>
461
462    <section title='Congestion Control'>
463
464      <t>The target bitrate of Opus can be adjusted at any point in time, thus
465      allowing efficient congestion control. Furthermore, the amount
466      of encoded speech or audio data encoded in a
467      single packet can be used for congestion control, since the transmission
468      rate is inversely proportional to the packet duration. A lower packet
469      transmission rate reduces the amount of header overhead, but at the same
470      time increases latency and loss sensitivity, so it ought to be used with
471      care.</t>
472
473      <t>Since UDP does not provide congestion control, applications that use
474      RTP over UDP SHOULD implement their own congestion control above the
475      UDP layer <xref target="RFC5405"/>. Work in the rmcat working group
476      <xref target="rmcat"/> describes the
477      interactions and conceptual interfaces necessary between the application
478      components that relate to congestion control, including the RTP layer,
479      the higher-level media codec control layer, and the lower-level
480      transport interface, as well as components dedicated to congestion
481      control functions.</t>
482    </section>
483
484    <section title='IANA Considerations'>
485      <t>One media subtype (audio/opus) has been defined and registered as
486      described in the following section.</t>
487
488      <section title='Opus Media Type Registration'>
489        <t>Media type registration is done according to <xref
490        target="RFC6838"/> and <xref target="RFC4855"/>.<vspace
491        blankLines='1'/></t>
492
493          <t>Type name: audio<vspace blankLines='1'/></t>
494          <t>Subtype name: opus<vspace blankLines='1'/></t>
495
496          <t>Required parameters:</t>
497          <t><list style="hanging">
498            <t hangText="rate:"> the RTP timestamp is incremented with a
499            48000 Hz clock rate for all modes of Opus and all sampling
500            rates. For data encoded with sampling rates other than 48000 Hz,
501            the sampling rate has to be adjusted to 48000 Hz.
502          </t>
503          </list></t>
504
505          <t>Optional parameters:</t>
506
507          <t><list style="hanging">
508            <t hangText="maxplaybackrate:">
509              a hint about the maximum output sampling rate that the receiver is
510              capable of rendering in Hz.
511              The decoder MUST be capable of decoding
512              any audio bandwidth but due to hardware limitations only signals
513              up to the specified sampling rate can be played back. Sending signals
514              with higher audio bandwidth results in higher than necessary network
515              usage and encoding complexity, so an encoder SHOULD NOT encode
516              frequencies above the audio bandwidth specified by maxplaybackrate.
517              This parameter can take any value between 8000 and 48000, although
518              commonly the value will match one of the Opus bandwidths
519              (<xref target="bandwidth_definitions"/>).
520              By default, the receiver is assumed to have no limitations, i.e. 48000.
521              <vspace blankLines='1'/>
522            </t>
523
524            <t hangText="sprop-maxcapturerate:">
525              a hint about the maximum input sampling rate that the sender is likely to produce.
526              This is not a guarantee that the sender will never send any higher bandwidth
527              (e.g. it could send a pre-recorded prompt that uses a higher bandwidth), but it
528              indicates to the receiver that frequencies above this maximum can safely be discarded.
529              This parameter is useful to avoid wasting receiver resources by operating the audio
530              processing pipeline (e.g. echo cancellation) at a higher rate than necessary.
531              This parameter can take any value between 8000 and 48000, although
532              commonly the value will match one of the Opus bandwidths
533              (<xref target="bandwidth_definitions"/>).
534              By default, the sender is assumed to have no limitations, i.e. 48000.
535              <vspace blankLines='1'/>
536            </t>
537
538            <t hangText="maxptime:"> the maximum duration of media represented
539            by a packet (according to Section&nbsp;6 of
540            <xref target="RFC4566"/>) that a decoder wants to receive, in
541            milliseconds rounded up to the next full integer value.
542            Possible values are 3, 5, 10, 20, 40, 60, or an arbitrary
543            multiple of an Opus frame size rounded up to the next full integer
544            value, up to a maximum value of 120, as
545            defined in <xref target='opus-rtp-payload-format'/>. If no value is
546              specified, the default is 120.
547              <vspace blankLines='1'/></t>
548
549            <t hangText="ptime:"> the preferred duration of media represented
550            by a packet (according to Section&nbsp;6 of
551            <xref target="RFC4566"/>) that a decoder wants to receive, in
552            milliseconds rounded up to the next full integer value.
553            Possible values are 3, 5, 10, 20, 40, 60, or an arbitrary
554            multiple of an Opus frame size rounded up to the next full integer
555            value, up to a maximum value of 120, as defined in <xref
556            target='opus-rtp-payload-format'/>. If no value is
557              specified, the default is 20.
558              <vspace blankLines='1'/></t>
559
560            <t hangText="maxaveragebitrate:"> specifies the maximum average
561            receive bitrate of a session in bits per second (b/s). The actual
562            value of the bitrate can vary, as it is dependent on the
563            characteristics of the media in a packet. Note that the maximum
564            average bitrate MAY be modified dynamically during a session. Any
565            positive integer is allowed, but values outside the range
566            6000 to 510000 SHOULD be ignored. If no value is specified, the
567            maximum value specified in <xref target='bitrate_by_bandwidth'/>
568            for the corresponding mode of Opus and corresponding maxplaybackrate
569            is the default.<vspace blankLines='1'/></t>
570
571            <t hangText="stereo:">
572              specifies whether the decoder prefers receiving stereo or mono signals.
573              Possible values are 1 and 0 where 1 specifies that stereo signals are preferred,
574              and 0 specifies that only mono signals are preferred.
575              Independent of the stereo parameter every receiver MUST be able to receive and
576              decode stereo signals but sending stereo signals to a receiver that signaled a
577              preference for mono signals may result in higher than necessary network
578              utilization and encoding complexity. If no value is specified,
579              the default is 0 (mono).<vspace blankLines='1'/>
580            </t>
581
582            <t hangText="sprop-stereo:">
583              specifies whether the sender is likely to produce stereo audio.
584              Possible values are 1 and 0, where 1 specifies that stereo signals are likely to
585              be sent, and 0 specifies that the sender will likely only send mono.
586              This is not a guarantee that the sender will never send stereo audio
587              (e.g. it could send a pre-recorded prompt that uses stereo), but it
588              indicates to the receiver that the received signal can be safely downmixed to mono.
589              This parameter is useful to avoid wasting receiver resources by operating the audio
590              processing pipeline (e.g. echo cancellation) in stereo when not necessary.
591              If no value is specified, the default is 0
592              (mono).<vspace blankLines='1'/>
593            </t>
594
595            <t hangText="cbr:">
596              specifies if the decoder prefers the use of a constant bitrate versus
597              variable bitrate. Possible values are 1 and 0, where 1 specifies constant
598              bitrate and 0 specifies variable bitrate. If no value is specified,
599              the default is 0 (vbr). When cbr is 1, the maximum average bitrate can still
600              change, e.g. to adapt to changing network conditions.<vspace blankLines='1'/>
601            </t>
602
603            <t hangText="useinbandfec:"> specifies that the decoder has the capability to
604            take advantage of the Opus in-band FEC. Possible values are 1 and 0.
605            Providing 0 when FEC cannot be used on the receiving side is
606            RECOMMENDED. If no
607            value is specified, useinbandfec is assumed to be 0.
608            This parameter is only a preference and the receiver MUST be able to process
609            packets that include FEC information, even if it means the FEC part is discarded.
610            <vspace blankLines='1'/></t>
611
612            <t hangText="usedtx:"> specifies if the decoder prefers the use of
613            DTX. Possible values are 1 and 0. If no value is specified, the
614            default is 0.<vspace blankLines='1'/></t>
615          </list></t>
616
617          <t>Encoding considerations:<vspace blankLines='1'/></t>
618          <t><list style="hanging">
619            <t>The Opus media type is framed and consists of binary data according
620            to Section&nbsp;4.8 in <xref target="RFC6838"/>.</t>
621          </list></t>
622
623          <t>Security considerations: </t>
624          <t><list style="hanging">
625            <t>See <xref target='security-considerations'/> of this document.</t>
626          </list></t>
627
628          <t>Interoperability considerations: none<vspace blankLines='1'/></t>
629	  <t>Published specification: RFC [XXXX]</t>
630	  <t>Note to the RFC Editor: Replace [XXXX] with the number of the published
631          RFC.<vspace blankLines='1'/></t>
632
633          <t>Applications that use this media type: </t>
634          <t><list style="hanging">
635            <t>Any application that requires the transport of
636            speech or audio data can use this media type. Some examples are,
637            but not limited to, audio and video conferencing, Voice over IP,
638            media streaming.</t>
639          </list></t>
640
641          <t>Fragment identifier considerations: N/A<vspace blankLines='1'/></t>
642
643          <t>Person &amp; email address to contact for further information:</t>
644          <t><list style="hanging">
645            <t>SILK Support silksupport@skype.net</t>
646            <t>Jean-Marc Valin jmvalin@jmvalin.ca</t>
647          </list></t>
648
649          <t>Intended usage: COMMON<vspace blankLines='1'/></t>
650
651          <t>Restrictions on usage:<vspace blankLines='1'/></t>
652
653          <t><list style="hanging">
654            <t>For transfer over RTP, the RTP payload format (<xref
655            target='opus-rtp-payload-format'/> of this document) SHALL be
656            used.</t>
657          </list></t>
658
659          <t>Author:</t>
660          <t><list style="hanging">
661            <t>Julian Spittka jspittka@gmail.com<vspace blankLines='1'/></t>
662            <t>Koen Vos koenvos74@gmail.com<vspace blankLines='1'/></t>
663            <t>Jean-Marc Valin jmvalin@jmvalin.ca<vspace blankLines='1'/></t>
664          </list></t>
665
666          <t> Change controller: IETF Payload Working Group delegated from the IESG</t>
667      </section>
668    </section>
669
670    <section title='SDP Considerations'>
671        <t>The information described in the media type specification has a
672        specific mapping to fields in the Session Description Protocol (SDP)
673        <xref target="RFC4566"/>, which is commonly used to describe RTP
674        sessions. When SDP is used to specify sessions employing Opus,
675        the mapping is as follows:</t>
676
677        <t>
678          <list style="symbols">
679            <t>The media type ("audio") goes in SDP "m=" as the media name.</t>
680
681            <t>The media subtype ("opus") goes in SDP "a=rtpmap" as the encoding
682            name. The RTP clock rate in "a=rtpmap" MUST be 48000 and the number of
683            channels MUST be 2.</t>
684
685            <t>The OPTIONAL media type parameters "ptime" and "maxptime" are
686            mapped to "a=ptime" and "a=maxptime" attributes, respectively, in the
687            SDP.</t>
688
689            <t>The OPTIONAL media type parameters "maxaveragebitrate",
690            "maxplaybackrate", "stereo", "cbr", "useinbandfec", and
691            "usedtx", when present, MUST be included in the "a=fmtp" attribute
692            in the SDP, expressed as a media type string in the form of a
693            semicolon-separated list of parameter=value pairs (e.g.,
694            maxplaybackrate=48000). They MUST NOT be specified in an
695            SSRC-specific "fmtp" source-level attribute (as defined in
696            Section&nbsp;6.3 of&nbsp;<xref target="RFC5576"/>).</t>
697
698            <t>The OPTIONAL media type parameters "sprop-maxcapturerate",
699            and "sprop-stereo" MAY be mapped to the "a=fmtp" SDP attribute by
700            copying them directly from the media type parameter string as part
701            of the semicolon-separated list of parameter=value pairs (e.g.,
702            sprop-stereo=1). These same OPTIONAL media type parameters MAY also
703            be specified using an SSRC-specific "fmtp" source-level attribute
704            as described in Section&nbsp;6.3 of&nbsp;<xref target="RFC5576"/>.
705            They MAY be specified in both places, in which case the parameter
706            in the source-level attribute overrides the one found on the
707            "a=fmtp" line. The value of any parameter which is not specified in
708            a source-level source attribute MUST be taken from the "a=fmtp"
709            line, if it is present there.</t>
710
711          </list>
712        </t>
713
714        <t>Below are some examples of SDP session descriptions for Opus:</t>
715
716        <t>Example 1: Standard mono session with 48000 Hz clock rate</t>
717          <figure>
718            <artwork>
719              <![CDATA[
720    m=audio 54312 RTP/AVP 101
721    a=rtpmap:101 opus/48000/2
722              ]]>
723            </artwork>
724          </figure>
725
726
727        <t>Example 2: 16000 Hz clock rate, maximum packet size of 40 ms,
728        recommended packet size of 40 ms, maximum average bitrate of 20000 bps,
729        prefers to receive stereo but only plans to send mono, FEC is desired,
730        DTX is not desired</t>
731
732        <figure>
733          <artwork>
734            <![CDATA[
735    m=audio 54312 RTP/AVP 101
736    a=rtpmap:101 opus/48000/2
737    a=fmtp:101 maxplaybackrate=16000; sprop-maxcapturerate=16000;
738    maxaveragebitrate=20000; stereo=1; useinbandfec=1; usedtx=0
739    a=ptime:40
740    a=maxptime:40
741            ]]>
742          </artwork>
743        </figure>
744
745        <t>Example 3: Two-way full-band stereo preferred</t>
746
747        <figure>
748          <artwork>
749            <![CDATA[
750    m=audio 54312 RTP/AVP 101
751    a=rtpmap:101 opus/48000/2
752    a=fmtp:101 stereo=1; sprop-stereo=1
753            ]]>
754          </artwork>
755        </figure>
756
757
758      <section title='SDP Offer/Answer Considerations'>
759
760          <t>When using the offer-answer procedure described in <xref
761          target="RFC3264"/> to negotiate the use of Opus, the following
762          considerations apply:</t>
763
764          <t><list style="symbols">
765
766            <t>Opus supports several clock rates. For signaling purposes only
767            the highest, i.e. 48000, is used. The actual clock rate of the
768            corresponding media is signaled inside the payload and is not
769            restricted by this payload format description. The decoder MUST be
770            capable of decoding every received clock rate. An example
771            is shown below:
772
773            <figure>
774              <artwork>
775                <![CDATA[
776    m=audio 54312 RTP/AVP 100
777    a=rtpmap:100 opus/48000/2
778                ]]>
779              </artwork>
780            </figure>
781            </t>
782
783            <t>The "ptime" and "maxptime" parameters are unidirectional
784            receive-only parameters and typically will not compromise
785            interoperability; however, some values might cause application
786            performance to suffer. <xref
787            target="RFC3264"/> defines the SDP offer-answer handling of the
788            "ptime" parameter. The "maxptime" parameter MUST be handled in the
789            same way.</t>
790
791            <t>
792              The "maxplaybackrate" parameter is a unidirectional receive-only
793              parameter that reflects limitations of the local receiver. When
794              sending to a single destination, a sender MUST NOT use an audio
795              bandwidth higher than necessary to make full use of audio sampled at
796              a sampling rate of "maxplaybackrate". Gateways or senders that
797              are sending the same encoded audio to multiple destinations
798              SHOULD NOT use an audio bandwidth higher than necessary to
799              represent audio sampled at "maxplaybackrate", as this would lead
800              to inefficient use of network resources.
801              The "maxplaybackrate" parameter does not
802              affect interoperability. Also, this parameter SHOULD NOT be used
803              to adjust the audio bandwidth as a function of the bitrate, as this
804              is the responsibility of the Opus encoder implementation.
805            </t>
806
807            <t>The "maxaveragebitrate" parameter is a unidirectional receive-only
808            parameter that reflects limitations of the local receiver. The sender
809            of the other side MUST NOT send with an average bitrate higher than
810            "maxaveragebitrate" as it might overload the network and/or
811            receiver. The "maxaveragebitrate" parameter typically will not
812            compromise interoperability; however, some values might cause
813            application performance to suffer, and ought to be set with
814            care.</t>
815
816            <t>The "sprop-maxcapturerate" and "sprop-stereo" parameters are
817            unidirectional sender-only parameters that reflect limitations of
818            the sender side.
819            They allow the receiver to set up a reduced-complexity audio
820            processing pipeline if the  sender is not planning to use the full
821            range of Opus's capabilities.
822            Neither "sprop-maxcapturerate" nor "sprop-stereo" affect
823            interoperability and the receiver MUST be capable of receiving any signal.
824            </t>
825
826            <t>
827              The "stereo" parameter is a unidirectional receive-only
828              parameter. When sending to a single destination, a sender MUST
829              NOT use stereo when "stereo" is 0. Gateways or senders that are
830              sending the same encoded audio to multiple destinations SHOULD
831              NOT use stereo when "stereo" is 0, as this would lead to
832              inefficient use of network resources. The "stereo" parameter does
833              not affect interoperability.
834            </t>
835
836            <t>
837              The "cbr" parameter is a unidirectional receive-only
838              parameter.
839            </t>
840
841            <t>The "useinbandfec" parameter is a unidirectional receive-only
842            parameter.</t>
843
844            <t>The "usedtx" parameter is a unidirectional receive-only
845            parameter.</t>
846
847            <t>Any unknown parameter in an offer MUST be ignored by the receiver
848            and MUST be removed from the answer.</t>
849
850          </list></t>
851
852        <t>
853	  The Opus parameters in an SDP Offer/Answer exchange are completely
854          orthogonal, and there is no relationship between the SDP Offer and
855          the Answer.
856        </t>
857      </section>
858
859      <section title='Declarative SDP Considerations for Opus'>
860
861        <t>For declarative use of SDP such as in Session Announcement Protocol
862        (SAP), <xref target="RFC2974"/>, and RTSP, <xref target="RFC2326"/>, for
863        Opus, the following needs to be considered:</t>
864
865        <t><list style="symbols">
866
867          <t>The values for "maxptime", "ptime", "maxplaybackrate", and
868          "maxaveragebitrate" ought to be selected carefully to ensure that a
869          reasonable performance can be achieved for the participants of a session.</t>
870
871          <t>
872            The values for "maxptime", "ptime", and of the payload
873            format configuration are recommendations by the decoding side to ensure
874            the best performance for the decoder.
875          </t>
876
877          <t>All other parameters of the payload format configuration are declarative
878          and a participant MUST use the configurations that are provided for
879          the session. More than one configuration can be provided if necessary
880          by declaring multiple RTP payload types; however, the number of types
881          ought to be kept small.</t>
882        </list></t>
883      </section>
884  </section>
885
886    <section title='Security Considerations' anchor='security-considerations'>
887
888      <t>Use of variable bitrate (VBR) is subject to the security considerations in
889      <xref target="RFC6562"/>.</t>
890
891      <t>RTP packets using the payload format defined in this specification
892      are subject to the security considerations discussed in the RTP
893      specification <xref target="RFC3550"/>, and in any applicable RTP profile such as
894      RTP/AVP <xref target="RFC3551"/>, RTP/AVPF <xref target="RFC4585"/>,
895      RTP/SAVP <xref target="RFC3711"/> or RTP/SAVPF <xref target="RFC5124"/>.
896      However, as "Securing the RTP Protocol Framework:
897      Why RTP Does Not Mandate a Single Media Security Solution"
898      <xref target="RFC7202"/> discusses, it is not an RTP payload
899      format's responsibility to discuss or mandate what solutions are used
900      to meet the basic security goals like confidentiality, integrity and
901      source authenticity for RTP in general.  This responsibility lays on
902      anyone using RTP in an application.  They can find guidance on
903      available security mechanisms and important considerations in Options
904      for Securing RTP Sessions [I-D.ietf-avtcore-rtp-security-options].
905      Applications SHOULD use one or more appropriate strong security
906      mechanisms.</t>
907
908      <t>This payload format and the Opus encoding do not exhibit any
909      significant non-uniformity in the receiver-end computational load and thus
910      are unlikely to pose a denial-of-service threat due to the receipt of
911      pathological datagrams.</t>
912    </section>
913
914    <section title='Acknowledgements'>
915    <t>Many people have made useful comments and suggestions contributing to this document.
916      In particular, we would like to thank
917      Tina le Grand, Cullen Jennings, Jonathan Lennox, Gregory Maxwell, Colin Perkins, Jan Skoglund,
918      Timothy B. Terriberry, Martin Thompson, Justin Uberti, Magnus Westerlund, and Mo Zanaty.</t>
919    </section>
920  </middle>
921
922  <back>
923    <references title="Normative References">
924      &rfc2119;
925      &rfc3389;
926      &rfc3550;
927      &rfc3711;
928      &rfc3551;
929      &rfc6838;
930      &rfc4855;
931      &rfc4566;
932      &rfc3264;
933      &rfc2326;
934      &rfc5576;
935      &rfc6562;
936      &rfc6716;
937    </references>
938
939    <references title="Informative References">
940      &rfc2974;
941      &rfc4585;
942      &rfc5124;
943      &rfc5405;
944      &rfc7202;
945
946      <reference anchor='rmcat' target='https://datatracker.ietf.org/wg/rmcat/documents/'>
947        <front>
948          <title>rmcat documents</title>
949          <author/>
950          <date/>
951          <abstract>
952            <t></t>
953          </abstract></front>
954      </reference>
955
956
957    </references>
958
959  </back>
960</rfc>
961