CT8022 Datasheet, PDF (19/194 Pages) List of Unclassifed Manufacturers – VOIP/VON G.723.1, G279AB TRUESPEECH CO-PROCESSOR

Version: 1.18
PRELIMINARY/CONFIDENTIAL
TrueSpeech® Co-Processor
3 Feature Overview
3.1 Introduction
The CT8022 is a full duplex TrueSpeech 8.5 and TrueSpeech G.723.1 real-time speech compression
Encoder/Decoder with built-in concurrent echo cancellation. It operates in conjunction with a Host processor and
one or two external A-to-D/D-to-A serial CODECs. The Host processor is responsible for managing the bi-
directional stream of compressed speech data provided by the CT8022. The CT8022 accepts uncompressed speech
in 8-bit A-law/µ-law or 16-bit Linear formats at 64/128 Kbits/sec from the receive channel of the external serial
CODEC, and compresses the speech to according to the Host-selected format. The compressed speech is passed in
frames every 30ms to the Host processor. As the CT8022 is compressing speech, it can concurrently accept
compressed speech from the Host for decompression, outputting the uncompressed speech to the transmit channel of
the external CODEC. The CT8022 supports independent and asynchronous operation of the Host receive and
transmit (compressed) data streams. The transmit and receive interfaces to the external A-to-D/D-to-A CODEC
operate synchronously. The device contains variable depth buffers for each data direction, and provides support of
frame deletion and insertion to accommodate sample clock differences between the source and destination devices.
The CT8022 can also operate in a CODEC-less mode as a speech compression/decompression accelerator device.
The CT8022 also provides auxiliary telephony functions such as DTMF generation and detection, and Call Progress
Tone generation and detection (e.g. dial tone). It is possible to configure the detectors and generators to operate on
either the input or the output audio streams (e.g. for detection on DTMF tones in the playback output audio channel).
3.2 Speech Modes
The CT8022 playback (transmit) and record (receive) channels can operate in several different speech modes:
• TrueSpeech G.723.1 6.3/5.3
• TrueSpeech 8.5 (see note 1)
• TrueSpeech 4.8 and 4.1, G.729 Annex A+B (see note 1)
• G.722 (see note 1)
• G.711 A-law/µ-law, 8-bit and 16-bit linear (uncompressed). It is possible to select the mode of operation of the
playback and record channel independently of each other.
Speech Coder
Used In
G.723.1
H.324 for video conferencing over dial-up V.34 Modems
6.3 & 5.3 KBPS
H.323 multi-media conferencing over LANs and TCP/IP type “packet-data” networks
(e.g. Internet)
TrueSpeech 8.5 (1)
8.5 KBPS
DSP Group speech coder built-in to Microsoft Windows 95. Also used in some
DSVD Modems and Internet Telephones
TrueSpeech 4.8 & 4.1
4.8/4.1 KBPS
DSP Group proprietary extension of the 6.3/5.3 coder for use in applications
requiring lower bit rates (e.g. proprietary non-standard extensions to H.324)
G.729A+B (1)
optional speech coder used in H.323
8 KBPS
G.711
Standard speech compression used in digital telephony systems (e.g. T1/E1)
µ-Law/A-Law
64 KBPS
G.722
7 KHz Wide bandwidth ADPCM audio coder used in H.320 (ISDN)
64 KBPS (1)
16-bit or 8-bit linear
Uncompressed audio (e.g. used in Microsoft WAVE files)
uncompressed data 128 or
64 KBPS
CT8022A11AQC FW Revision 0118
DSP GROUP, INC., 3120 SCOTT BOULEVARD
19
SANTA CLARA, CA 95054 PH: 408 986 – 4300 FAX: 408 986 – 4490
All specifications are subject to change without prior notice.