NSAM265SR Datasheet, PDF (8/58 Pages) National Semiconductor (TI) – CompactSPEECH Digital Speech Processors

1 0 Theory of Operation (Continued)
1 7 3 Memory Operating Modes in NSAM265SF
The NSAM265SF supports two operating modes of the
FLASH memory manager (selected by the CFG command)
Normal Mode
In this mode the NSAM265SF always keeps one free block
for memory management The EV MEMFULL event is set
when all but one good AFLASH blocks are full and record-
ing on the last available free block is not allowed
Maximum recording time in this mode
On one 8-Mbit FLASH (no bad blocks) device 28 minutes
and 8 seconds
On one 4-Mbit FLASH (no bad blocks) device 13 minutes
and 8 seconds
Memory Intensive Mode
All good blocks are available for message recording As
long as there is one free block available for memory man-
agement this mode is the same as normal mode The Com-
pactSPEECH sets the EV MEMLOW event when approxi-
mately 20 seconds recording time remain on the one-be-
fore-last FLASH block The EV MEMFULL event is set
only when all the blocks are full
However if the last good block is used for recording the
NSAM265SF no longer has a free block for memory man-
agement and therefore may fail to free memory space i e
execute the FR command The microcontroller must delete
all messages with the MESSAGE SAFE bit cleared to
guarantee that at least one block is erased and can be
used for memory management
When there is no block available for memory management
and you attempt to record a message with the MESSAGE
SAFE bit set the CompactSPEECH sets the EV
MEMFULL event and does not allow you to record Thus it
is possible to avoid a situation where a safe message must
be deleted to free a FLASH block
Maximum recording time in this mode
On one 8 Mbits FLASH (no bad blocks) device 30 minutes
On one 4 Mbits FLASH (no bad blocks) device 15 minutes
1 8 TONE AND NO-ENERGY DETECTION
The CompactSPEECH detects DTMF busy and dial tones
and no energy (VOX) This enables remote control opera-
tions and call progress Detection is active throughout the
operation of the CompactSPEECH Detection can be con-
figured using the SDET (Set Detectors Mask) command
which controls the reporting of occurrence of tones and the
RDET (Reset Detectors) command which resets the detec-
tors
DTMF
DTMF detection may be reported at starting point ending
point or both The report is made through the status word
(for further details see GSW command in Section 3 2)
The DTMF detector specifications as measured on the line
input using the NSV-AM265-DAA board are summarized be-
low (see Table 1-2)
Echo Cancellation
Echo cancellation is a technique used to improve the per-
formance of DTMF tone detection during speech synthesis
tone generation and OGM playback For echo cancellation
to work properly AGC must not be active in parallel Thus
to take advantage of echo cancellation the microcontroller
must control the AGC i e disable the AGC during PLAY
SYNTHESIS and TONE GENERATE states and enable it
again afterwards If AGC can not be disabled do not use
echo cancellation The microcontroller should use the CFG
command to activate deactivate echo cancellation (For fur-
ther details see Section 3 2 )
Echo cancellation applies only to DTMF tones Busy and
dial tones detection is not affected by this technique
TABLE 1-2 DTMF Detector Specifications
Play
Detection Sensitivity b33 with Echo Canceller
(Note A)
b23 w o Echo Canceller
Accepted
DTMF Length
l_{50 ms}
Record
Idle
b40 dBm
l_{40 ms}
Frequency Tolerance
g1 5%
g1 5%
S N Ratio
Minimum Spacing
(Note B)
Normal Twist
12 dB
l_{45 ms}
8 dB
12 dB
l_{45 ms}
8 dB
Reverse Twist
(Note C)
4 dB or 8 dB
4 dB or 8 dB
Note A Performance depends on DAA design
Note B If the interval between two consecutive DTMF tones is less than or
equal to 20 ms the two are detected as one long DTMF tone If the interval
between two consecutive DTMF tones is between 20 ms and 45 ms sepa-
rate detection is unpredictable
Note C Determined according to the DTMF REV TWIST tunable param-
eter value
8