Group PJMEDIA_CONFIG

group PJMEDIA_CONFIG

Some compile time configuration settings.

Defines

PJMEDIA_CONF_USE_SWITCH_BOARD

Specify whether we prefer to use audio switch board rather than conference bridge.

Audio switch board is a kind of simplified version of conference bridge, but not really the subset of conference bridge. It has stricter rules on audio routing among the pjmedia ports and has no audio mixing capability. The power of it is it could work with encoded audio frames where conference brigde couldn’t.

Default: 0

PJMEDIA_HAS_LEGACY_SOUND_API

This macro has been deprecated in releasee 1.1. Please see http://trac.pjsip.org/repos/wiki/Audio_Dev_API for more information. This macro has been deprecated in releasee 1.1. Please see http://trac.pjsip.org/repos/wiki/Audio_Dev_API for more information. This macro controls whether the legacy sound device API is to be implemented, for applications that still use the old sound device API (sound.h). If this macro is set to non-zero, the sound_legacy.c will be included in the compilation. The sound_legacy.c is an implementation of old sound device (sound.h) using the new Audio Device API.

Please see http://trac.pjsip.org/repos/wiki/Audio_Dev_API for more info.

PJMEDIA_SND_DEFAULT_REC_LATENCY

Specify default sound device latency, in milisecond.

PJMEDIA_SND_DEFAULT_PLAY_LATENCY

Specify default sound device latency, in milisecond.

Default is 160ms for Windows Mobile and 140ms for other platforms.

PJMEDIA_WSOLA_IMP_NULL

This denotes implementation of WSOLA using null algorithm. Expansion will generate zero frames, and compression will just discard some samples from the input.

This type of implementation may be used as it requires the least processing power.

PJMEDIA_WSOLA_IMP_WSOLA

This denotes implementation of WSOLA using fixed or floating point WSOLA algorithm. This implementation provides the best quality of the result, at the expense of one frame delay and intensive processing power requirement.

PJMEDIA_WSOLA_IMP_WSOLA_LITE

This denotes implementation of WSOLA algorithm with faster waveform similarity calculation. This implementation provides fair quality of the result with the main advantage of low processing power requirement.

PJMEDIA_WSOLA_IMP

Specify type of Waveform based Similarity Overlap and Add (WSOLA) backend implementation to be used. WSOLA is an algorithm to expand and/or compress audio frames without changing the pitch, and used by the delaybuf and as PLC backend algorithm.

Default is PJMEDIA_WSOLA_IMP_WSOLA

PJMEDIA_WSOLA_MAX_EXPAND_MSEC

Specify the default maximum duration of synthetic audio that is generated by WSOLA. This value should be long enough to cover burst of packet losses. but not too long, because as the duration increases the quality would degrade considerably.

Note that this limit is only applied when fading is enabled in the WSOLA session.

Default: 80

PJMEDIA_WSOLA_TEMPLATE_LENGTH_MSEC

Specify WSOLA template length, in milliseconds. The longer the template, the smoother signal to be generated at the expense of more computation needed, since the algorithm will have to compare more samples to find the most similar pitch.

Default: 5

PJMEDIA_WSOLA_DELAY_MSEC

Specify WSOLA algorithm delay, in milliseconds. The algorithm delay is used to merge synthetic samples with real samples in the transition between real to synthetic and vice versa. The longer the delay, the smoother signal to be generated, at the expense of longer latency and a slighty more computation.

Default: 5

PJMEDIA_WSOLA_PLC_NO_FADING

Set this to non-zero to disable fade-out/in effect in the PLC when it instructs WSOLA to generate synthetic frames. The use of fading may or may not improve the quality of audio, depending on the nature of packet loss and the type of audio input (e.g. speech vs music). Disabling fading also implicitly remove the maximum limit of synthetic audio samples generated by WSOLA (see PJMEDIA_WSOLA_MAX_EXPAND_MSEC).

Default: 0

PJMEDIA_MAX_PLC_DURATION_MSEC

Limit the number of calls by stream to the PLC to generate synthetic frames to this duration. If packets are still lost after this maximum duration, silence will be generated by the stream instead. Since the PLC normally should have its own limit on the maximum duration of synthetic frames to be generated (for PJMEDIA’s PLC, the limit is PJMEDIA_WSOLA_MAX_EXPAND_MSEC), we can set this value to a large number to give additional flexibility should the PLC wants to do something clever with the lost frames.

Default: 240 ms

PJMEDIA_SOUND_BUFFER_COUNT

Specify number of sound buffers. Larger number is better for sound stability and to accommodate sound devices that are unable to send frames in timely manner, however it would probably cause more audio delay (and definitely will take more memory). One individual buffer is normally 10ms or 20 ms long, depending on ptime settings (samples_per_frame value).

The setting here currently is used by the conference bridge, the splitter combiner port, and dsound.c.

Default: (PJMEDIA_SND_DEFAULT_PLAY_LATENCY+20)/20

PJMEDIA_HAS_ALAW_ULAW_TABLE

Specify which A-law/U-law conversion algorithm to use. By default the conversion algorithm uses A-law/U-law table which gives the best performance, at the expense of 33 KBytes of static data. If this option is disabled, a smaller but slower algorithm will be used.

PJMEDIA_HAS_G711_CODEC

Unless specified otherwise, G711 codec is included by default.

PJMEDIA_RESAMPLE_NONE

No resampling.

PJMEDIA_RESAMPLE_LIBRESAMPLE

Sample rate conversion using libresample.

PJMEDIA_RESAMPLE_SPEEX

Sample rate conversion using Speex.

PJMEDIA_RESAMPLE_LIBSAMPLERATE

Sample rate conversion using libsamplerate (a.k.a Secret Rabbit Code)

PJMEDIA_RESAMPLE_IMP

Select which resample implementation to use. Currently pjmedia supports:

Default is PJMEDIA_RESAMPLE_LIBRESAMPLE

PJMEDIA_FILE_PORT_BUFSIZE

Specify whether libsamplerate, when used, should be linked statically into the application. This option is only useful for Visual Studio projects, and when this static linking is enabled Default file player/writer buffer size.

PJMEDIA_MAX_FRAME_DURATION_MS

Maximum frame duration (in msec) to be supported. This (among other thing) will affect the size of buffers to be allocated for outgoing packets.

PJMEDIA_MAX_MTU

Max packet size to support.

PJMEDIA_DTMF_DURATION

DTMF/telephone-event duration, in timestamp.

PJMEDIA_RTP_NAT_PROBATION_CNT

Number of RTP packets received from different source IP address from the remote address required to make the stream switch transmission to the source address.

PJMEDIA_RTCP_NAT_PROBATION_CNT

Number of RTCP packets received from different source IP address from the remote address required to make the stream switch RTCP transmission to the source address.

PJMEDIA_ADVERTISE_RTCP

Specify whether RTCP should be advertised in SDP. This setting would affect whether RTCP candidate will be added in SDP when ICE is used. Application might want to disable RTCP advertisement in SDP to reduce the message size.

Default: 1 (yes)

PJMEDIA_RTCP_INTERVAL

Interval to send RTCP packets, in msec

PJMEDIA_RTCP_IGNORE_FIRST_PACKETS

Tell RTCP to ignore the first N packets when calculating the jitter statistics. From experimentation, the first few packets (25 or so) have relatively big jitter, possibly because during this time, the program is also busy setting up the signaling, so they make the average jitter big.

Default: 25.

PJMEDIA_RTCP_STAT_HAS_RAW_JITTER

Specify whether RTCP statistics includes raw jitter statistics. Raw jitter is defined as absolute value of network transit time difference of two consecutive packets; refering to “difference D” term in interarrival jitter calculation in RFC 3550 section 6.4.1.

Default: 0 (no).

PJMEDIA_RTCP_NORMALIZE_FACTOR

Specify the factor with wich RTCP RTT statistics should be normalized if exceptionally high. For e.g. mobile networks with potentially large fluctuations, this might be unwanted.

Use (0) to disable this feature.

Default: 3.

PJMEDIA_RTCP_STAT_HAS_IPDV

Specify whether RTCP statistics includes IP Delay Variation statistics. IPDV is defined as network transit time difference of two consecutive packets. The IPDV statistic can be useful to inspect clock skew existance and level, e.g: when the IPDV mean values were stable in positive numbers, then the remote clock (used in sending RTP packets) is faster than local system clock. Ideally, the IPDV mean values are always equal to 0.

Default: 0 (no).

PJMEDIA_HAS_RTCP_XR

Specify whether RTCP XR support should be built into PJMEDIA. Disabling this feature will reduce footprint slightly. Note that even when this setting is enabled, RTCP XR processing will only be performed in stream if it is enabled on run-time on per stream basis. See PJMEDIA_STREAM_ENABLE_XR setting for more info.

Default: 0 (no).

PJMEDIA_STREAM_ENABLE_XR

The RTCP XR feature is activated and used by stream if enable_rtcp_xr field of pjmedia_stream_info structure is non-zero. This setting controls the default value of this field.

Default: 0 (disabled)

PJMEDIA_RTCP_RX_SDES_BUF_LEN

Specify the buffer length for storing any received RTCP SDES text in a stream session. Usually RTCP contains only the mandatory SDES field, i.e: CNAME.

Default: 64 bytes.

PJMEDIA_STREAM_VAD_SUSPEND_MSEC

Specify how long (in miliseconds) the stream should suspend the silence detector/voice activity detector (VAD) during the initial period of the session. This feature is useful to open bindings in all NAT routers between local and remote endpoint since most NATs do not allow incoming packet to get in before local endpoint sends outgoing packets.

Specify zero to disable this feature.

Default: 600 msec (which gives good probability that some RTP packets will reach the destination, but without filling up the jitter buffer on the remote end).

PJMEDIA_STREAM_CHECK_RTP_PT

Perform RTP payload type checking in the stream. Normally the peer MUST send RTP with payload type as we specified in our SDP. Certain agents may not be able to follow this hence the only way to have communication is to disable this check.

Default: 1

PJMEDIA_CODEC_MAX_SILENCE_PERIOD

Specify the maximum duration of silence period in the codec, in msec. This is useful for example to keep NAT binding open in the firewall and to prevent server from disconnecting the call because no RTP packet is received.

This only applies to codecs that use PJMEDIA’s VAD (pretty much everything including iLBC, except Speex, which has its own DTX mechanism).

Use (-1) to disable this feature.

Default: 5000 ms

PJMEDIA_SILENCE_DET_THRESHOLD

Suggested or default threshold to be set for fixed silence detection or as starting threshold for adaptive silence detection. The threshold has the range from zero to 0xFFFF.

PJMEDIA_SILENCE_DET_MAX_THRESHOLD

Maximum silence threshold in the silence detector. The silence detector will not cut the audio transmission if the audio level is above this level.

Use 0x10000 (or greater) to disable this feature.

Default: 0x10000 (disabled)

PJMEDIA_HAS_SPEEX_AEC

Speex Accoustic Echo Cancellation (AEC). By default is enabled.

PJMEDIA_SDP_NEG_PREFER_REMOTE_CODEC_ORDER

This specifies the behavior of the SDP negotiator when responding to an offer, whether it should rather use the codec preference as set by remote, or should it rather use the codec preference as specified by local endpoint.

For example, suppose incoming call has codec order “8 0 3”, while local codec order is “3 0 8”. If remote codec order is preferable, the selected codec will be 8, while if local codec order is preferable, the selected codec will be 3.

If set to non-zero, the negotiator will use the codec order as specified by remote in the offer.

Note that this behavior can be changed during run-time by calling pjmedia_sdp_neg_set_prefer_remote_codec_order().

Default is 1 (to maintain backward compatibility)

PJMEDIA_HAS_RTCP_IN_SDP

Support for sending and decoding RTCP port in SDP (RFC 3605). Default is equal to PJMEDIA_ADVERTISE_RTCP setting.

PJMEDIA_ADD_RTPMAP_FOR_STATIC_PT

This macro controls whether pjmedia should include SDP rtpmap attribute for static payload types. SDP rtpmap for static payload types are optional, although they are normally included for interoperability reason.

Note that there is also a run-time variable to turn this setting on or off, defined in endpoint.c. To access this variable, use the following construct

   extern pj_bool_t pjmedia_add_rtpmap_for_static_pt;

   // Do not include rtpmap for static payload types (<96)
   pjmedia_add_rtpmap_for_static_pt = PJ_FALSE;

Default: 1 (yes)

PJMEDIA_RTP_PT_TELEPHONE_EVENTS

This macro declares the payload type for telephone-event that is advertised by PJMEDIA for outgoing SDP. If this macro is set to zero, telephone events would not be advertised nor supported.

If this value is changed to other number, please update the PJMEDIA_RTP_PT_TELEPHONE_EVENTS_STR too.

PJMEDIA_RTP_PT_TELEPHONE_EVENTS_STR

Macro to get the string representation of the telephone-event payload type.

PJMEDIA_TONEGEN_MAX_DIGITS

Maximum tones/digits that can be enqueued in the tone generator.

PJMEDIA_TONEGEN_SINE

The math’s sine(), floating point. This has very good precision but it’s the slowest and requires floating point support and linking with the math library.

PJMEDIA_TONEGEN_FLOATING_POINT

Floating point approximation of sine(). This has relatively good precision and much faster than plain sine(), but it requires floating- point support and linking with the math library.

PJMEDIA_TONEGEN_FIXED_POINT_CORDIC

Fixed point using sine signal generated by Cordic algorithm. This algorithm can be tuned to provide balance between precision and performance by tuning the PJMEDIA_TONEGEN_FIXED_POINT_CORDIC_LOOP setting, and may be suitable for platforms that lack floating-point support.

PJMEDIA_TONEGEN_FAST_FIXED_POINT

Fast fixed point using some approximation to generate sine waves. The tone generated by this algorithm is not very precise, however the algorithm is very fast.

PJMEDIA_TONEGEN_ALG

Specify the tone generator algorithm to be used. Please see http://trac.pjsip.org/repos/wiki/Tone_Generator for the performance analysis results of the various tone generator algorithms.

Default value:

  • PJMEDIA_TONEGEN_FLOATING_POINT when PJ_HAS_FLOATING_POINT is set

  • PJMEDIA_TONEGEN_FIXED_POINT_CORDIC when PJ_HAS_FLOATING_POINT is not set

PJMEDIA_TONEGEN_FIXED_POINT_CORDIC_LOOP

Specify the number of calculation loops to generate the tone, when PJMEDIA_TONEGEN_FIXED_POINT_CORDIC algorithm is used. With more calculation loops, the tone signal gets more precise, but this will add more processing.

Valid values are 1 to 28.

Default value: 10

PJMEDIA_TONEGEN_FADE_IN_TIME

Enable high quality of tone generation, the better quality will cost more CPU load. This is only applied to floating point enabled machines.

By default it is enabled when PJ_HAS_FLOATING_POINT is set.

This macro has been deprecated in version 1.0-rc3. Fade-in duration for the tone, in milliseconds. Set to zero to disable this feature.

Default: 1 (msec)

PJMEDIA_TONEGEN_FADE_OUT_TIME

Fade-out duration for the tone, in milliseconds. Set to zero to disable this feature.

Default: 2 (msec)

PJMEDIA_TONEGEN_VOLUME

The default tone generator amplitude (1-32767).

Default value: 12288

PJMEDIA_HANDLE_G722_MPEG_BUG

Enable support for SRTP media transport. This will require linking with libsrtp from the third_party directory.

By default it is enabled. Enable support to handle codecs with inconsistent clock rate between clock rate in SDP/RTP & the clock rate that is actually used. This happens for example with G.722 and MPEG audio codecs. See:

  • G.722 : RFC 3551 4.5.2

  • MPEG audio : RFC 3551 4.5.13 & RFC 3119

Also when this feature is enabled, some handling will be performed to deal with clock rate incompatibilities of some phones.

By default it is enabled.

PJMEDIA_TRANSPORT_SPECIFIC_INFO_MAXCNT

Transport info (pjmedia_transport_info) contains a socket info and list of transport specific info, since transports can be chained together (for example, SRTP transport uses UDP transport as the underlying transport). This constant specifies maximum number of transport specific infos that can be held in a transport info.

PJMEDIA_TRANSPORT_SPECIFIC_INFO_MAXSIZE

Maximum size in bytes of storage buffer of a transport specific info.

PJMEDIA_STREAM_KA_EMPTY_RTP

Value to be specified in PJMEDIA_STREAM_ENABLE_KA setting. This indicates that an empty RTP packet should be used as the keep-alive packet.

PJMEDIA_STREAM_KA_USER

Value to be specified in PJMEDIA_STREAM_ENABLE_KA setting. This indicates that a user defined packet should be used as the keep-alive packet. The content of the user-defined packet is specified by PJMEDIA_STREAM_KA_USER_PKT. Default content is a CR-LF packet.

PJMEDIA_STREAM_KA_USER_PKT

The content of the user defined keep-alive packet. The format of the packet is initializer to pj_str_t structure. Note that the content may contain NULL character.

PJMEDIA_STREAM_KA_INTERVAL

Specify another type of keep-alive and NAT hole punching mechanism (the other type is PJMEDIA_STREAM_VAD_SUSPEND_MSEC and PJMEDIA_CODEC_MAX_SILENCE_PERIOD) to be used by stream. When this feature is enabled, the stream will initially transmit one packet to punch a hole in NAT, and periodically transmit keep-alive packets.

When this alternative keep-alive mechanism is used, application may disable the other keep-alive mechanisms, i.e: by setting PJMEDIA_STREAM_VAD_SUSPEND_MSEC to zero and PJMEDIA_CODEC_MAX_SILENCE_PERIOD to -1.

The value of this macro specifies the type of packet used for the keep-alive mechanism. Valid values are PJMEDIA_STREAM_KA_EMPTY_RTP and PJMEDIA_STREAM_KA_USER.

The duration of the keep-alive interval further can be set with PJMEDIA_STREAM_KA_INTERVAL setting.

Default: 0 (disabled) Specify the keep-alive interval of PJMEDIA_STREAM_ENABLE_KA mechanism, in seconds.

Default: 5 seconds

PJMEDIA_JBUF_DISC_MIN_GAP

Minimum gap between two consecutive discards in jitter buffer, in milliseconds.

Default: 200 ms

PJMEDIA_JBUF_PRO_DISC_MIN_BURST

Minimum burst level reference used for calculating discard duration in jitter buffer progressive discard algorithm, in frames.

Default: 1 frame

PJMEDIA_JBUF_PRO_DISC_MAX_BURST

Maximum burst level reference used for calculating discard duration in jitter buffer progressive discard algorithm, in frames.

Default: 200 frames

PJMEDIA_JBUF_PRO_DISC_T1

Duration for progressive discard algotithm in jitter buffer to discard an excessive frame when burst is equal to or lower than PJMEDIA_JBUF_PRO_DISC_MIN_BURST, in milliseconds.

Default: 2000 ms

PJMEDIA_JBUF_PRO_DISC_T2

Duration for progressive discard algotithm in jitter buffer to discard an excessive frame when burst is equal to or greater than PJMEDIA_JBUF_PRO_DISC_MAX_BURST, in milliseconds.

Default: 10000 ms