TL EE12378
NSAM265SRNSAM265SF
CompactSPEECH
Digital
Speech
Processors
PRELIMINARY
November 1995
NSAM265SR NSAM265SF
CompactSPEECH
TM
Digital Speech Processors
General Description
The NSAM265SR and the NSAM265SF are members of
National
Semiconductor's
CompactSPEECH
Digital
Speech processors family These processors provide Digital
Answering Machine (DAM) functionality to embedded sys-
tems Both processors are based on the NSAM265
Unless specified otherwise all references to the Compact-
SPEECH processor in this document apply to both the
NSAM265SR and the NSAM265SF
The CompactSPEECH processor integrates the functions of
a traditional Digital Signal Processing (DSP) chip and a 16-
bit CompactRISC
TM
embedded Risc processor core The
device contains system support functions such as DRAM
Controller
Interrupt Control Unit
Codec Interface
MI-
CROWlRE
TM
interface WATCHDOG
TM
timer and a Clock
Generator
The CompactSPEECH processor operates as a slave pe-
ripheral that is controlled by an external microcontroller via
a serial MICROWIRE interface In a typical DAM environ-
ment the microcontroller controls the analog circuits but-
tons and display and activates the CompactSPEECH by
sending it commands The CompactSPEECH processor ex-
ecutes the commands and returns status information to the
microcontroller
The CompactSPEECH firmware implements voice compres-
sion and decompression tone detection and generation
message storage management on-chip speech synthesis
for time and day stamp and support for user-defined voice
prompts in various languages
The NSAM265SR CompactSPEECH supports DRAM
ARAM for message storage while the NSAM265SF sup-
ports FLASH AFLASH In all other respects the processors
are identical
The CompactSPEECH implements echo cancellation tech-
niques to support improved DTMF tone detection during
message playback
The CompactSPEECH supports speech synthesis the tech-
nology used to create voice prompts from predefined words
and phrases stored in a vocabulary
The CompactSPEECH can synthesize messages in various
languages in addition to the on-chip English vocabulary via
the International Vocabulary Support (IVS) mechanism Syn-
thesized messages can be stored on an external ROM One
ROM can contain several vocabularies in various lan-
guages The NSAM265SF can also store vocabularies on
FLASH memory DAM manufacturers can thus create ma-
chines that ``speak'' in different languages simply by using
other vocabularies For more details about IVS refer to the
IVS User's Manual
Features
Y
Designed around National's 16-bit CompactRISC proc-
essor
Y
16-bit architecture and implementation
Y
20 48 MHz operation
Y
On-chip DSP Module (DSPM) for high speed DSP
operations
Y
On-chip Codec clock generation and interface
Y
Power-down mode
Y
MICROWIRE interface to an external microcontroller
Y
Storage and management of messages
Y
Programmable message tag for message categoriza-
tion e g
Mailboxes InComing Messages (ICM) Out-
Going Messages (OGM)
Y
Skip forward or backward during message playback
Y
Variable speed playback
Y
Built-in vocabulary for speech synthesis and support
for external vocabularies using expansion ROM
Y
Multi-lingual speech synthesis using International Vo-
cabulary Support (IVS)
Y
DTMF and single tone generation and detection
Y
DTMF tone detection during OutGoing Message play-
back
Y
Telephone line functions including busy and dial tone
detection
Y
Real-time clock
Y
Direct access to message memory
Y
Supports long-frame and short-frame codecs
Y
Available in PLCC 68-pin and PQFP 100-pin packages
NSAM265SR only
Y
On-chip ARAM DRAM Controller for 4-Mbit (1M x 4)
and 16-Mbit (4M x 4) devices
Y
15 minutes recording on a 4-Mbit ARAM
Y
Supports various ARAM configurations No glue logic
required
Y
Storage of up to 1600 messages
Y
Production diagnostics support
NSAM265SF only
Y
Supports
4-Mbit
and
8-Mbit
byte
wide
FLASH
AFLASH devices
Y
Up to 15 minutes recording on a 4-Mbit FLASH
Y
Supports various AFLASH configurations No glue logic
required for a single AFLASH configuration
Y
The number of messages that can be stored is limited
only by memory size
Y
Supports prerecorded IVS and OGM on FLASH
TRI-STATE
is a registered trademark of National Semiconductor Corporation
CompactSPEECH
TM
CompactRISC
TM
MICROWIRE
TM
and WATCHDOG
TM
are trademarks of National Semiconductor Corporation
C1995 National Semiconductor Corporation
RRD-B30M115 Printed in U S A
Block Diagrams
NSAM265SF Basic Configuration
TL EE 12378 1
NSAM265SR Basic Configuration
TL EE 12378 2
2
Table of Contents
1 0 THEORY OF OPERATION
1 1 Overview
1 2 The State Machine
1 3 Command Execution
1 4 Tunable Parameters
1 5 Messages
1 5 1 Message Tag
1 6 ARAM Support
1 7 FLASH Support
1 7 1 Block Erasure
1 7 2 FLASH Endurance
1 7 3 Memory Operating Modes in NSAM265SF
1 8 Tone and No-Energy Detection
1 9 Speech Synthesis
1 9 1 Explanation of Terms
1 9 2 Internal Vocabulary
1 9 3 External (International) Vocabularies
1 10 Initialization
2 0 FUNCTIONAL DESCRIPTION
2 1 Introduction
2 2 Resetting
2 3 The Serial Microwire Interface
2 3 1 Signal Description
2 3 2 Signal Use in the Interface Protocol
2 3 3 Interface Protocol Error Handling
2 4 CODEC Interface
2 5 DRAM ARAM Access (NSAM265SR)
2 5 1 Refreshing a DRAM ARAM
2 6 DRAM ARAM Specifications (NSAM265SR)
2 7 DRAM ARAM Configurations (NSAM265SR)
2 8 FLASH AFLASH Access (NSAM265SF)
2 9 FLASH AFLASH Specifications
2 10 FLASH AFLASH Configurations
2 11 IVS Access
2 12 Clocking
2 13 Power-Down Mode
2 14 Power and Grounding
3 0 COMMAND SET
3 1 Commands Summary
3 2 Commands Description
APPENDIX A
DEVICE SPECIFICATIONS
A 1 Pin Assignment
A 2 Absolute Maximum Ratings
A 3 Electrical Characteristics
A 4 Switching Characteristics
APPENDIX B
SCHEMATIC DIAGRAMS
3
1 0 Theory of Operation
1 1 OVERVIEW
The CompactSPEECH is a digital speech processor which
provides Digital Answering Machine (DAM) functionality to
embedded systems such as fax machines or stand-alone
answering machines
The CompactSPEECH processor is based on a powerful
16-bit RISC CPU with Digital Signal Processing (DSP) mod-
ule
However
since the CompactSPEECH firmware is
masked in the internal ROM it requires neither 16-bit nor
DSP programming
The CompactSPEECH processor is designed to operate as
a slave peripheral that is controlled by an external micro-
controller via a simple MICROWIRE based serial link The
microcontroller is responsible for system-level control (e g
buttons display ring detection) while the CompactSPEECH
is responsible for speech and tone operations such as re-
cording playback tone detection and speech synthesis
The NSAM265SR CompactSPEECH supports DRAM and
Audio grade DRAM (ARAM) as a storage device ARAMs
are cheaper than DRAMs which reduces the total system
cost
The NSAM265SR CompactSPEECH has an on-chip DRAM
controller which reduces the chip count for complete DAM
functionality to three (the CompactSPEECH ARAM and a
codec) The CompactSPEECH supports various ARAM con-
figurations and automatically recognizes the actual configu-
ration
The NSAM265SF CompactSPEECH supports FLASH and
Audio grade FLASH (AFLASH) as a storage device A DAM
design which incorporates FLASH technology does not re-
quire a battery backup for the message storage device thus
reducing the system cost and complexity
When necessary the CompactSPEECH can be switched to
power-down mode which considerably reduces the power
consumption of the whole system
The CompactSPEECH includes an on-chip English vocabu-
lary for time-and-day stamp announcements and voice
prompts In addition to the on-chip vocabulary the Com-
pactSPEECH supports external vocabularies (resident on
external ROM or AFLASH) which can be used to implement
voice prompts in various languages Several vocabularies
can be supported simultaneously
This chapter describes the features of the Compact-
SPEECH processor and how they work together
1 2 THE STATE MACHINE
The CompactSPEECH functions as a state machine It
changes state either in response to a command sent by the
microcontroller after execution of the command is complet-
ed or as a result of an internal event (e g memory full or
power failure)
The CompactSPEECH may be in one of the following
states
RESET
The CompactSPEECH is initialized to this state after a full
hardware reset by the RESET signal (see Section 2 2)
CompactSPEECH detectors (VOX call progress tones and
DTMF tones) are not active In all other states the detec-
tors are active (See the SDET and RDET commands for
further details )
IDLE
This is the state from which most commands are executed
As soon as a command and all its parameters are received
the CompactSPEECH starts executing the command
PLAY
In this state a message is decompressed and played
RECORD
In this state a message is compressed and recorded into
the message memory
SYNTHESIS
An individual word or a sentence is synthesized from the on-
chip or an external vocabulary
TONE
GENERATE
The CompactSPEECH generates single or DTMF tones
MEMORY
READ
The CompactSPEECH reads a 32-byte block from the mes-
sage memory and sends it to the external microcontroller
MEMORY
WRITE
The CompactSPEECH accepts a 32-byte block from the ex-
ternal microcontroller and writes it to the message memory
MEMORY
FREE
The CompactSPEECH takes memory space that was freed
by the DM and DMS commands and makes it available for
new messages This process occurs only with NSAM265SF
where FLASH memory is used for message storage
After receiving an asynchronous command (see Section
1 3) such as P (Playback) R (Record) SW (Say Words) or
GT (Generate Tone) the CompactSPEECH switches to the
appropriate state and executes the command until it is com-
pleted or an S (Stop) or PA (Pause) command is received
from the microcontroller
When an asynchronous command execution is completed
the CompactSPEECH switches to the IDLE state
Table 1-1 shows the CompactSPEECH commands the
source states in which these commands are valid and the
result states which the CompactSPEECH enters as a result
of the command
4
1 0 Theory of Operation
(Continued)
TABLE 1-1 CompactSPEECH States and Transitions
Command
Description
Source State
Result State
Configuration and Status Commands
AMAP
Check and Map ARAM
IDLE
IDLE
CFG
Configure CompactSPEECH
RESET
RESET
CVOC
Check Vocabulary
IDLE
IDLE
GEW
Get Error Word
All states
No change
FR
Free Memory
IDLE
MEMORY
FREE
GCFG
Get Configuration
RESET IDLE
No change
GI
Get Information
PLAY RECORD SYNTHESIS
No change
TONE
GENERATE IDLE
MEMORY
FREE
GMS
Get Memory Status
IDLE
IDLE
GSW
Get Status Word
All states
No change
GTD
Get Time and Day
IDLE
IDLE
INIT
Initialize System
RESET IDLE
IDLE
INJ
Inject IVS Data
IDLE
IDLE
MR
Memory Reset
IDLE
IDLE
PDM
Go to Power-Down Mode
IDLE
IDLE
RDET
Reset Detectors
MEMORY
FREE IDLE
No change
S
Stop
All states but RESET
IDLE
SDET
Set Detectors Mask
IDLE MEMORY
FREE
No change
SETD
Set Time and Day
IDLE MEMORY
FREE
No change
SV
Set Vocabulary Type
IDLE
IDLE
TUNE
Tune
IDLE MEMORY
FREE
No change
Speech Commands
AMSG
Append to Current Message
IDLE
RECORD
GT
Generate Tone
IDLE
TONE
GENERATE
P
Playback
IDLE
PLAY
PA
Pause
PLAY RECORD SYNTHESIS
No change
TONE
GENERATE IDLE
R
Record
IDLE
RECORD
RES
Resume
PLAY RECORD SYNTHESIS
No change
TONE
GENERATE IDLE
SAS
Say Argumented Sentence
IDLE
SYNTHESIS
SB
Skip Backward
PLAY IDLE
No change
SE
Skip to End of Message
PLAY IDLE
No change
SF
Skip Forward
PLAY IDLE
No change
SO
Say One Word
IDLE
SYNTHESIS
SPS
Set Playback Speed
PLAY SYNTHESIS
No change
IDLE MEMORY
FREE
SS
Say Sentence
IDLE
SYNTHESIS
SW
Say Words
IDLE
SYNTHESIS
VC
Volume Control
PLAY SYNTHESIS IDLE
No change
TONE
GENERATE
5
PDF 
ZIP