English
Language : 

CT8022 Datasheet, PDF (23/194 Pages) List of Unclassifed Manufacturers – VOIP/VON G.723.1, G279AB TRUESPEECH CO-PROCESSOR
Version: 1.18
PRELIMINARY/CONFIDENTIAL
TrueSpeech® Co-Processor
3.18 Tone Detection
The CT8022 contains a set of four independent programmable call progress tone filters which can be used to detect
call progress tones such as dial tone, busy tone, ringing (or ringback) tone and fax calling tone (CNG). The tone
detector can be configured to operate on either the incoming (record) data stream or outgoing (playback) data
stream. Tone detection can be performed either on the output or on the input speech directions.
3.19 DTMF Detection
The CT8022 includes a DTMF detector that can detect all 16 DTMF digits including the digits ABCD. The tone
detector can be configured to operate on either the incoming (record) data stream or outgoing (playback) data
stream. DTMF detection can be performed either in the output or in the input speech directions. This allows
detection of DTMF digits in the output audio stream.
3.20 Full Duplex Speakerphone
The CT8022 can operate in either standalone speakerphone or in DSVD speakerphone mode (for use in a
videoconferencing system). In DSVD speakerphone mode, the CT8022 provides Acoustical Echo Cancellation
between the speaker and microphone connection, concurrent with speech compression operation. In standalone full-
duplex speakerphone mode, Acoustical Echo Cancellation as well as near-end (electrical) echo cancellation to the
telephone line connection are provided. Standalone speakerphone mode provides support for creating an analog
telephone line-based speakerphone.
Unlike many speakerphone implementations, the CT8022 implements a full-duplex speakerphone. In most
speakerphones, only one party at a time can speak and, the other party is always muted, sometimes by as much as 50
dB or more. This is half-duplex speakerphone operation. In half-duplex operation a speakerphone listens to both
the microphone and the telephone line inputs and tries to detect which side is speaking. It allows input from the
speaking side and mutes input from the other side. If this were not done, acoustical feedback or howling would take
place. The major problem with a half-duplex speakerphone is that a user experiences break-up during the
conversation when the speakerphone incorrectly identifies the speaking party (due to noise) or both parties attempt to
talk at the same time.
The CT8022 implements a full-duplex speakerphone. This implementation allows both parties to speak and to be
heard simultaneously. It does this using DSP technology called Acoustical Echo Cancellation.
Acoustical echoing occurs when the speakerphone microphone picks up the audio signal from the speakerphone
speaker, echoing the output (speaker) signal back along the telephone line. The signal transmitted through the
telephone line is also echoed by it and output to the speaker again. This creates a feedback loop in which the signal
from the telephone line is output from the speaker, picked up by the microphone, sent to the telephone line, echoed
by the telephone line and re-output by the speaker. This causes howling or howl round, a loud annoying high-
pitched noise. In the CT8022, howling is prevented by the use of echo cancellation. This is an adaptive filtering
process where the CT8022 compensates for the speaker-to-microphone echo and subtracts the outgoing speaker
signal from the incoming microphone signal, reducing or eliminating the acoustic echo. Refer to Section 9.21 for
more details on Speakerphone operation.
3.21 Acoustical Echo Cancellation
The CT8022 includes an Acoustical Echo Canceller (AEC) as part of its full-duplex speakerphone feature. In
addition to its use in standalone speakerphone mode, the AEC can be operated concurrently with the CT8022
full-duplex speech mode (DSVD mode). This enables the CT8022 to provide acoustic echo cancellation between the
speaker and microphone while simultaneously providing full duplex speech compression/decompression. This
makes possible hands-free DSVD (DSVD or Video Conferencing Speakerphone) applications avoiding use of a
headset or handset for speech input/output. The AEC implementation included in the CT8022 contains features that
allow the echo canceller to be used to cancel non-acoustical echoes such as the electrical signal echo from a 2 to 4-
wire telephone hybrid. This option supports use of the CT8022 in a digital speech telephony gateway-type
application.
CT8022A11AQC FW Revision 0118
DSP GROUP, INC., 3120 SCOTT BOULEVARD
23
SANTA CLARA, CA 95054 PH: 408 986 – 4300 FAX: 408 986 – 4490
All specifications are subject to change without prior notice.