Loading...

MonteCarlo API

PIKA MonteCarlo SDK supports both DSP-based and HMP (host-based) media processing technology. Developers can choose between these two architectures, depending on what is most suitable for the application. Both approaches can work for TDM, IP or mixed TDM/IP environments.

PIKA’s HMP technology incorporates interfaces for both IP and TDM networks, allows applications to make and receive calls and includes media processing applications including play and record, tone and DTMF detection/generation, conference summing and low latency switching. Codec support includes G.711 and G.729. All the standard voice, tone, and modem applications that have traditionally run only on DSPs are available on the host without loss of quality or performance.

The following table provides a comparison of features supported in the PIKA SDK for DSP and HMP boards.

Feature

DSP

HMP

Audio

Yes

Yes

Play

Yes

Yes

Record

Yes

Yes

Formats:
raw
µ-law
A-Law
GSM
8-bit PCM
16-bit PCM
3-bit ADPCM
4-bit OKI ADPCM
4-bit Dialogic ADPCM (VOX)

Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes

Yes
Yes
Yes
Yes
No
Yes
No
Yes
Yes

VAD (Voice Activity Detection)

Yes

Yes

Audio Gain Pad

Yes

No

Automatic Gain Control (AGC)

Yes

Yes

CAS (RBS, MFR2, MFR1)

Yes

Yes

Call Progress/Analysis

No

Yes (with GrandPrix API)

Conferencing

Yes

Yes

Gain Pad

Yes

Yes

Summing

Yes

Yes

Switching

Yes

Yes

DTMF Clamping

Yes

Yes

Echo Cancellation

Yes

Yes

FAX G3 (T.30)

Yes

Yes

FSK

Yes

Yes

HDLC

Yes

Yes

OpenVPOS Custom Apps

Yes

No

Pulse Detection / Generation

Yes

No

Tones
Generation

Yes

Yes

Detection

Yes

Yes

DTMF

Yes

Yes

VoIP
SIP

Yes

Yes

RTP

Yes

Yes

RTP + RFC2833

No

Yes

RTP By-pass switching option

No

Yes

G.711

Yes

Yes

G.726

Yes

No

G.729 a/b

No

Yes

Detailed technical descriptions of each application follow. You can also read the Application Papers by following this link.

 


Audio (DSP and HMP – some exceptions)

The audio application supports the following audio formats, to be used for playback and recording.

OKI/PIKA ADPCM 4-bit 8kHz
OKI/PIKA ADPCM 3-bit 4kHz, 6kHz, 8kHz, 11kHz (DSP only)
OKI/PIKA ADPCM 4-bit 4kHz, 6kHz, 8kHz, 11kHz
µ-Law / A-Law / Linear PCM 8-bit 4kHz, 6kHz, 8kHz, 11kHz (DSP only)
µ-Law / A-Law / Linear PCM 16-bit 4kHz, 6kHz, 8kHz, 11kHz

Typical audio applications

  • Voicemail systems
  • Auto attendant systems, and
  • Audio logging.
  • Forward to speech recognition apps

Features of the audio application include:

  • Support for audio with variable buffer sizes (including small buffers) is a feature of our play/record DSP/HMP application. Flexible audio buffers allow you to set the buffer size so that you can optimize your system for your unique needs.

Voice activity detection (VAD), a feature of audio, is based on energy detection and provides time stamping of the state changes between the presence of voice activity (start event) and the removal of voice activity (stop events). The PIKA VAD API allows you to configure the maximum threshold (level to start recording); minimum threshold (level to stop recording); activation debounce time; deactivation debounce time; and pre-speech buffer size.

Voice Activity Detection

A feature of the VAD application is audio time stamping. You can time-stamp an audio file for easier retrieval.

Automatic Gain Control (AGC) is another feature of the audio application that is available for both record and play. AGC changes the level of the speech signal to be closer to a desired level. This is accomplished without subjectively degrading the audio quality of the signal.

Typical applications include:

  • Voicemail
  • Auto attendants
  • PC-PBXs

Application Paper on Audio (DSP)


Caller ID Detection (DSP and HMP)

Caller ID is a feature of Frequency Shift Key (FSK)


CAS (DSP and HMP)

T1 RBS
PIKA Technologies’ CAS implementation supports the following standard T1 CAS protocols: E&M Wink Start (ANSI TI.403), Loop Start (ANSI TI.403), and Ground Start (ANSI TI.403). There is also a mechanism to customize your own protocol or any non-default supported protocol.

E1 MFR2
Support for digital channel associated signaling (CAS) protocols now includes System R2, used widely in many parts of the world. PIKA Technologies’ CAS implementation supports the following standard E1 R2 protocol variants: Cuba , and Korea . Like the T1 CAS protocols, the E1 R2 protocols may be customized to meet the needs of any R2 protocol country variants.

Application Paper for CAS (DSP)
Application Paper for MFR2 (DSP)


Conferencing (DSP and HMP)

Sample summing algorithm (DSP only) mixes all the inputs into the conference bridge so that each of the participants hears the sum of all the other inputs, except their own. The output for each participant is then determined by subtracting the participant input signal from the conference sum.

Sample switching (switching) (DSP only) identifies the principal speaker and mutes all other input channels. In this scenario, there is a single active connection between all participants with the system connecting the loudest speaker’s channel, at any given time, to the conference bridge.

Advanced conferencing (HMP only) is a further enhancement to Loudest talker that allows N-loudest talker to input in the conference and be heard.

  • Suitable for building very large conferences
  • Capable of supporting N active talkers, where N can be any number from one to the number of conferees in the conference
  • Fast DTMF clamping
  • Detecting and reporting active speakers
  • Stability enhancements, allowing large volume gains if desired
  • Automatic Gain Control
  • Can be programmed to have behavior similar to sample switching and sample summation
  • Low latency. Programmable latency with the lowest latency equal to the basic mode conference latency. Adding about 14 ms of additional latency is required to completely clamp all DTMF digits

The conferencing application allows multiple audio sources to be conferenced together on one card or across multiple cards.

Typical applications include:

  • 3-way calling
  • large-scale conference calls

Application Paper on Voice Conferencing (DSP)


DTMF Detection & Generation (DSP and HMP)

The DTMF detection & generation application decodes both standard and non-standard DTMF tones that a telephone produces.

Typical applications include:

  • Voicemail
  • Auto attendant
  • PC-PBX
  • DTMF logging

Echo Cancellation (DSP and HMP)

The echo cancellation application greatly reduces or eliminates the echo that is sometimes produced during calls that occur over great distances, especially over satellites.

Typical applications include:

  • Speech/voice recognition
  • Long distance phone calls (both North American and overseas
  • Voice over IP (VoIP Internet-based phone calls)
  • Voice activated dialing

Pause/resume is a feature of echo cancellation. A user no longer needs to wait for a tone played event before recording. This application improves performance, capability, and memory efficiency of the solution.

Application Paper for Echo Cancellation (DSP)


Fax (transmit only / transmit and receive) (DSP and HMP)

The FAX application will allow reception and/or transmission of standard facsimile images thereby enhancing our messaging communications functionality offerings.

Typical applications include:

  • Fax broadcasting
  • Fax-on-demand
  • Real-time fax
  • Store and forward fax
  • Interactive fax-back
  • Fax mailboxes
  • Unified messaging
  • Call centers

Application Paper on Fax (HMP)


Generic FSK Modem (DSP and HMP)

 

The Frequency Shift Key (FSK) application allows reception and transmission of standard modem signals, as well as custom signals, at speeds of up to 2400 bps.

Caller identification uses FSK. The caller ID detector decodes the North American calling number identification (CNID) signals that are transmitted between the first and second ring on an incoming call. Caller ID service provides the following call information:

  • Date and time of the call
  • Number of the calling party
  • Directory information (name and possible address)
  • Error messages such as blocked number or out of area number
  • Typical caller ID applications include:
  • Customer information screen pops, and
  • Caller ID logging.
  • Modems: V.21, V.23, Bell 102, Bell 203, Caller ID NA, Caller ID Japan, Caller ID Europe, MBI MSK

GSM 6.10 Audio Compression (DSP and HMP)

PIKA supports GSM 6.10 audio compression. The GSM codec enables playing and recording audio in Microsoft Windows applications. The MM-series voice cards support both standard GSM and Microsoft GSM codec 6.10.

GSM has the following main features:

  • Two different methods for encoding:
    – Standard GSM, as described in the ETSI standard
    – Microsoft GSM, compatible with MS Windows utilities
  • Loaded onto the DSP as a special DSP SRE file through the OpenVPOS interface
  • Controlled via the existing Audio API
  • Programmable input (record) gain
  • Programmable output (play) gain
  • Record AGC
  • Play AGC
  • Internet mode of operation (Standard GSM only)
    – Support for dynamic jitter buffer management – Skip and insert frame
    – Lost frame reconstruction
  • Playback pause / unpause
  • VAD
    – Speech pre-buffering

OpenVPOS / Custom Feature Development (DSP only)

PIKA’s proprietary voice processing operating system (VPOS) has been opened enabling the development of custom applications writen by a developer or by PIKA. Although we offer OpenVPOS, you may prefer to contract the custom application development project to our experienced PIKA Custom Engineering Services team. If you have an algorithm that is CPU intensive, our team can help.

An example of an application developed through OpenVPOS is GSM support on DSP boards.


Pulse Detection & Generation (DSP only)

The pulse detection application translates the ‘clicking’ or audio pulses from older rotary dial phones into the equivalent DTMF tones. This allows users with older phone equipment to access telephony applications written only for DTMF phones.

Typical applications include:

  • Voicemail
  • Auto attendant
  • PC-PBXs

Application Paper for Dial Pulse Detection (DSP)


SIP

Session Initiation Protocol (SIP) is a protocol used to set up and tear down sessions that allow media (such as voice data) to be exchanged by devices connected on an IP network. SIP allows devices to negotiate how the media is transported, but is not used to transport the media.

 


Speech Detector (DSP and HMP)

The speech detector application can be used to detect the presence of speech. A more common use however is to detect the absence of speech. A programmable timer in the speech detector can be used to report a timeout if a DSP port has not detected any speech activity.

Typical applications include:

  • Voicemail
  • Auto attendant
  • PC-PBX

Application Paper for Speech Detector (DSP)


Tone Detection & Generation (DSP and HMP)

The tone detection & generation application allows the detection & generation of both DTMF and custom tones for use in DTMF dialing, ring back tones, and dial tone.

Typical applications include:

  • Voicemail
  • Auto attendants
  • PC-PBXs

Application Paper on Tone Generation (DSP)
Application Paper on Advanced Tone Detection (ATD) (DSP)
Application Paper on DTMF (DSP)


RTP Features (DSP and HMP – some exceptions)

The real-time protocol (RTP) application is a block process with a variable block size. This number depends on the block size of the codec process that the RTP interacts with and it can change dynamically. It provides end-to-end transport functions suitable for applications transmitting interactive real-time data, such as audio.

  • Supports G.711 (PCMA and PCMU)
  • G.726 (40, 32, 24 and 16 Kbps) (DSP only)
  • In HMP environments, supports G.729a/b (HMP only)
  • Supports 1ms frame resolution for G.711 and G.726 reception
  • Supports dynamic payload types
  • Supports enable / disable out-of-band DTMF signaling as specified in RFC2833 (HMP only)
  • Supports fixed delay and adaptive jitter buffer management
  • Creates data for RTCP reports
  • Disables DTMF detection while generating a digit on the receiver side in order to prevent false detection of echoed DTMF signals
  • Flexible architecture. A number of parameters are visible to the user and if necessary, they can be altered to adjust the performance of the application:
  • Changeable jitter buffer size
  • Packetization rate: The number of encoder frames per RTP packet
  • Initial Latency: This parameter indicates the number of frames to be placed in the jitter buffer before starting the decoder
  • Play tone delay: delays the generation of a DTMF tone when the RFC 2833 protocol is being used. Delaying the start of a tone ensures that tone generation does not end prematurely when an RFC 2833 RTP packet is late or lost (HMP only)
  • RTP By-pass switching (HMP only)
  • Echo cancellation, up to 128 ms tail length, tested against the G.168 spec

Application Paper on RTP (DSP)


G.711 Features (DSP and HMP)

 

  • Fully compliant with ITU-T G.711 recommendation
  • Optimized implementation
  • Configurable input and output gains
  • Supported by the PIKA VoIP RTP application
  • Recovers lost or late frames using audio reconstruction
  • Supports skipping and insertion of frames (remove this line)
  • Supports out of band DTMF (RFC 2833)

Application Paper on G.711 (DSP)


G.726 Features (DSP only)

  • Fully bit exact with ITU-T G.726 Annex A
  • Optimized implementation
  • Configurable coding rate
  • Configurable input and output gains
  • Recovers lost or late frames using audio reconstruction

Application Paper for G.726 (DSP)


G.729a and G.729b (HMP only)

  • Fully bit exact with ITU-T G.729 Annex A and B
  • Optimized implementation
  • Configurable input and output gains
  • Recovers lost or late frames using audio reconstruction
  • Supported by the PIKA VoIP RTP application
  • Supports out of band DTMF (RFC 28331)