RSC-164I [ETC]

General Purpose Microcontroller Featuring Speech Recognition, Speaker Verification, and Speech Synthesis.; 通用微控制器配备语音识别,说话人确认和语音合成。
RSC-164I
型号: RSC-164I
厂家: ETC    ETC
描述:

General Purpose Microcontroller Featuring Speech Recognition, Speaker Verification, and Speech Synthesis.
通用微控制器配备语音识别,说话人确认和语音合成。

微控制器 语音合成
文件: 总5页 (文件大小:63K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
RSC-164i  
Recognition · Synthesis · Control  
General Purpose Microcontroller Featuring Speech Recognition,  
Speaker Verification, and Speech Synthesis.  
GENERAL DESCRIPTION  
FEATURES  
The RSC-164i, from the Interactive Speech™ family of  
products, is a low-cost microcontroller designed for use in  
consumer electronics. The RSC-164i combines an 8-bit  
microcontroller with high-quality speaker-independent  
and speaker-dependent speech recognition, speaker  
verification, and speech synthesis. Products can use one  
or all of the RSC-164i features in a single application.  
Full Range of Speech Capabilities  
· Speaker-independent speech recognition  
· Speaker-dependent speech recognition  
· High quality speech synthesis and sound effects  
· Speaker verification  
Integrated Single-Chip Solution  
The RSC-164i employs a sophisticated neural network  
that learns to classify sound data. On-chip speech  
recognition algorithms reach an accuracy of greater than  
96% for speaker-independent recognition and greater than  
99% for speaker-dependent recognition. Sensory’s neural  
network approach (patent pending) eliminates the need for  
expensive signal processing or extensive RAM storage.  
· 4 MIPS 8-bit microcontroller  
· On-chip A/D and D/A converters, digital filtering  
· 32kHz clock for time keeping  
· Internal 64kbytes ROM; 384 bytes RAM  
· 12 general purpose I/O lines  
· On-chip output PWM for direct speaker drive  
The highly-integrated nature of the chip reduces external  
parts count. A complete system may be built with few  
additional parts other than a battery, speaker, microphone,  
and audio input support circuitry. The RSC-164i is  
similar to the RSC-164, but does not support a parallel  
memory interface. All program, synthesis, recognition,  
and verification code must fit in the on-chip ROM.  
Low Power Requirements  
· 3.5-5.0V supply  
· ~10mA operating  
RSC-164i Block Diagram  
Oscillator  
External  
Analog Input  
Preamp  
PWM  
DAC  
RAM  
ROM  
Microphone  
ADC  
Speaker  
Microcontroller  
Digital Logic  
AGC  
RSC-164i  
General  
Purpose I/O  
From the Interactive Speech™ Line of Products  
RSC-164i  
DATA SHEET  
RSC-164i OVERVIEW  
POWER  
The RSC-164i features  
a
high-performance 8-bit  
The typical operating current is 10 mA operating at  
14.32 MHz.  
microcontroller with on-chip A/D, D/A, RAM and ROM.  
The RSC-164i is designed to bring a high degree of  
integration and versatility into low-cost, power-sensitive  
consumer applications.  
RSC-164i Architecture Diagram  
AIN0  
ADC  
Various functional units have been integrated onto the  
CPU core in order to reduce total system cost and  
increase system reliability without degrading system  
performance. The RSC-164i delivers 4 MIPS of integer  
performance at 14.32 MHz providing maximum  
performance at minimum cost.  
MUX  
AIN1  
SH  
ADC  
DACOUT  
DAC  
ANALOG  
CONTROL  
REGISTER SPACE  
384 bytes  
BUFOUT  
/PWM  
The CPU core embedded in the RSC-164i is an 8-bit,  
variable-length-instruction, microcontroller. The instruc-  
tion set is loosely based on Intel’s 8051Ô , having a  
variety of addressing mode mov instructions. But the  
RSC-164i processor avoids the limitations of dedicated  
A, B, and DPTR registers by having completely  
symmetrical source and destinations for all instructions.  
The 384 bytes of internal RAM are organized as a  
Register Space.  
PULSE WIDTH  
MODULATOR  
STACK SPACE  
8 bytes  
XI1, XO1  
OSC1  
CPU  
TIMER1  
TIMER2  
INTERNAL ROM  
32K x 8  
HIGH  
LOW  
XI2, XO2  
32K x 8  
OSC2  
-RESET  
-TE1/  
TIMING AND CONTROL  
PWM  
P0.0-P0.7  
SPEECH RECOGNITION  
BREAK POINT  
REGISTER  
The RSC-164i uses a neural network to perform speaker-  
independent or speaker-dependent speech recognition.  
Speaker-dependent recognition requires external serial  
memory to store speech recognition templates. Speaker-  
independent recognition requires on-chip ROM to store  
the words to be recognized. The RSC-164i has several  
additional speech recognition features as described  
below.  
P1.0-P1.3  
RSC-164i ARCHITECTURE  
The RSC-164i is a highly integrated device that  
combines:  
Continuous listening allows the chip to continuously  
listen for a specific word. With this feature a product can  
be used in a normal environment and only “activates”  
when a specific word, preceded by quiet, is spoken.  
·
·
·
8-bit microcontroller  
On-chip ROM (64 kbytes) and RAM (384 bytes)  
A/D converter and D/A converter  
SPEECH SYNTHESIS  
Two bi-directional ports provide 12 general purpose I/O  
pins to communicate with external devices. The RSC-  
164i has a high frequency (14.32 MHz) oscillator as well  
as a low frequency (32,768 Hz) oscillator suitable for  
timekeeping applications. The processor clock can be  
selected from either source, with a selectable divider  
value. The device performs speech recognition when  
running at 14.32 MHz. There are two programmable 8-  
bit counters / timers, one derived from each oscillator.  
The RSC-164i provides high-quality speech synthesis by  
using a hybrid of a time-domain compression scheme  
that improves on conventional ADPCM and a customized  
reuse of sounds. Speech synthesis uses on-chip ROM to  
store audio sounds for synthesis. The RSC-164i supports  
approximately 25 seconds of speech.  
SPEAKER VERIFICATION  
The RSC-164i can perform text-dependent speaker  
verification. After a speaker trains the chip on a specific  
word, the chip will be able to identify whether that word  
is spoken by the original speaker.  
From the Interactive Speech™ Line of Products  
DATA SHEET  
RSC-164i  
A microphone with an external preamp converts sound  
into an audio signal that is fed to the RSC-164i. The gain  
of the external preamp may be controlled by the RSC-  
164i by using two of the I/O lines. The RSC-164i uses an  
ADC (Analog-to-Digital Converter) to convert incoming  
analog speech signals into digital data. The output audio  
signal of the RSC-164i is derived from a DAC (Digital-  
to-Analog Converter) or PWM (Pulse Width Modulator).  
14.32 MHz (3.5V-5.0V)  
Pins XI2 and XO2  
32768 Hz (3.5V-5.0V)  
Oscillator #2  
Oscillator #1 works with an external crystal, a ceramic  
resonator or LC. Use of Oscillator #2 requires a crystal  
for precision timing.  
CLOCK  
The RSC-164i uses a fully static core – the processor can  
be stopped (by removing the clock source) and restarted  
without causing a reset or losing contents of internal  
registers. Static operation is functional from DC to 14.32  
MHz.  
USING THE RSC-164i  
Creating applications using the RSC-164i requires the  
development of electronic circuitry, software code, and  
speech/music data files. Software code for the RSC-164i  
can be developed by Sensory or by external programmers  
using the RSC Development Kit. For more information  
about development tools and services, please contact  
Sensory. A typical product will require about $0.80 -  
$1.50 (in high volume) of additional components, in  
addition to the RSC-164i.  
Typically the processor clock runs from a 14.32 MHz  
crystal with no divisor and one wait state. This creates  
internal RAM cycles of 70 nsec duration and internal  
ROM of 140 nsec duration.  
TIMERS/COUNTERS  
The two independent oscillators of the RSC-164i provide  
counts to two internal timers. Each of the two timers  
consists of an 8-bit reload value register and an 8-bit up-  
counter. The reload register is readable and writeable by  
the processor.  
RSC-164i INSTRUCTION SET  
The instruction set for the RSC-164i has 52 instructions  
comprising 8 move, 7 rotate, 11 branch, 11 register  
arithmetic, 9 immediate arithmetic, and 6 miscellaneous  
instructions. All instructions are 3 bytes or fewer, and no  
instruction requires more than 8 clock cycles to execute.  
INTERRUPTS  
GENERAL PURPOSE I/O  
The RSC-164i allows for five interrupt sources, as  
selected by software. Each has its own mask bit and  
request bit in the IMR and IRQ registers respectively.  
The following events can generate interrupts:  
The RSC-164i has 12 general purpose I/O pins (P0.0-  
P0.7, P1.0-P1.3). Each pin can be programmed as an  
input with weak pull-up (~400kW equivalent device);  
input with strong pull-up (~10kW equivalent device);  
input without pull-up, or as an output. This is  
accomplished by having 24 bits of configuration registers  
for the I/O pins (Port Control Register A and Port  
Control Register B for ports 0 and 1).  
·
·
·
·
·
Positive edge on Port 0, bit 0  
Overflow of Timer 1  
Overflow of Timer 2  
Sensory reserved functions  
Completion of PWM sample period  
EXTERNAL MEMORY  
ANALOG OUTPUT  
The RSC-164i can connect serially through two I/O lines  
to a serial EEPROM for applications with low data  
storage and low access speed requirements. Speaker-  
dependent and speaker verification can therefore be  
performed on the RSC-164i with such external memory.  
The RSC-164i offers two separate options for analog  
output. The DAC (Digital to Analog Converter) output  
provides a general purpose 10-bit analog output that may  
be used for speech output (with the inclusion of an audio  
amplifier), or other purposes requiring an analog  
waveform. For speech applications that require driving a  
small speaker, the PWM (Pulse-Width Modulator) output  
can be used instead of the DAC output. The PWM  
output can directly drive a 32 ohm speaker.  
OSCILLATORS  
Two independent oscillators in the RSC-164i provide a  
high-frequency clock and a 32kHz time-keeping clock.  
The oscillator characteristics are as follows:  
Oscillator #1:  
Pins XI1, XO1  
From the Interactive Speech™ Line of Products  
RSC-164i  
DATA SHEET  
ABSOLUTE MAXIMUM RATINGS  
Any pin to GND  
-0.1V to +7.5V  
WARNING: Stressing the RSC-164i beyond the  
Operating temperature (TO)  
Soldering temperature  
Power dissipation  
0°C to +70°C  
260°C for 10 sec  
TBD  
0°C to +70°C;  
VDD=3.5-5.0V;  
VSS=0V  
“Absolute Maximum Ratings” may cause permanent  
damage. These are stress ratings only. Operation  
beyond the “Operating Conditions” is not  
recommended and extended exposure beyond the  
“Operating Conditions” may affect device  
reliability.  
Operating Conditions  
THE INTERACTIVE SPEECH™ PRODUCT LINE  
The Interactive Speech line of ICs was developed to “bring life to products” through advanced speech and audio technology.  
These chips allow products to think, talk, hear and play music. The Interactive Speech chips were designed for consumer  
telephony products and cost-sensitive consumer electronic applications such as home electronics, personal security, and  
personal communication. The product line includes general purpose microcontrollers (RSC-164, RSC-164i, RSC-132i) and  
application specific standard speech ASSPs (Voice Password™ and Voice Direct™).  
RSC-164  
The RSC-164 is a low-cost 8-bit microcontroller designed for use in consumer electronics. It is a  
fully integrated microcontroller and includes A/D, D/A, ROM, and RAM circuitry on chip. The  
RSC-164 can perform a full range of speech/audio functions including speech recognition,  
speaker verification, speech and music synthesis, and voice record/playback.  
RSC-132i  
The RSC-132i, the lowest cost member of the RSC series, is a speech enabled microcontroller designed specifically for the toy  
industry. The RSC-132i can support various combinations of speech technologies. The RSC-132i combines an 8-bit  
microcontroller with high quality speaker-independent and speaker-dependent speech recognition, speech synthesis, and  
speaker verification. This chips has 32 kBytes of ROM and limited I/O pins  
Voice Password™ ASSP  
The Voice Passwordä ASSP provides consumer products with low cost biometric security. The chip lets products “lock out”  
unauthorized access by verifying key words and/or voices. Using text dependent speaker verification technology, Voice  
Password can secure products at a variety of security thresholds for many different applications.  
Voice Direct™ ASSP  
The Voice Directä ASSP provides cost-sensitive products with speaker-dependent speech recognition, speech synthesis and  
DTMF tone generation. This easy-to-use, pin-configurable chip requires no custom programming and can recognize up to 60  
trained words. The Voice Directä ASSP is most ideal for consumer telephony products which feature voice dialing.  
IMPORTANT NOTICES  
Sensory reserves the right to make changes to or to discontinue any product or service identified in this publication at any time without notice in order to improve design and supply the best  
possible product. Sensory does not assume responsibility for use of any circuitry other than circuitry entirely embodied in a Sensory product. Information contained herein is provided  
gratuitously and without liability to any user. Reasonable efforts have been made to verify the accuracy of this information but no guarantee whatsoever is given as to the accuracy or as to its  
applicability to particular uses.  
Applications described in this data sheet are for illustrative purposes only, and Sensory makes no warranties or representations that the RSC series of products will be suitable for such  
applications. In every instance, it must be the responsibility of the user to determine the suitability of the products for each application. Sensory products are not authorized for use as critical  
components in life support devices or systems.  
Sensory conveys no license or title, either expressed or implied, under any patent, copyright, or mask work right to the RSC series of products, and Sensory makes balance between  
recognition and synthesis no warranties or representations that the RSC series of products are free from patent, copyright, or mask work right infringement, unless otherwise specified.  
© 1996 SENSORY, INC.  
ALL RIGHTS RESERVED  
P/N 80-0024-3  
5 2 1 East Weddell Drive  
Sunnyvale, CA 94089  
TEL: (408) 744-9000  
FAX: (408) 744-1299  
Sensory is registered by the U.S. Patent and Trademark Office.  
All other trademarks or registered trademarks are the property  
of their respective owners.  
From the Interactive Speech™ Line of Products  
DATA SHEET  
RSC-164i  
Nothing contained herein shall be construed as a recommendation to use any product in violation of existing patents or other rights of third parties. The sale of any Sensory product is subject  
to all Sensory Terms and Conditions of Sales and Sales Policies.  
From the Interactive Speech™ Line of Products  

相关型号:

RSC-300

Speech Recognition Microcontroller
ETC

RSC-300DIE

Speech Recognition Microcontroller
ETC

RSC-300DWF

Speech Recognition Microcontroller
ETC

RSC-300QFP

Speech Recognition Microcontroller
ETC

RSC-364

Speech Recognition Microcontroller
ETC

RSC-364DIE

Speech Recognition Microcontroller
ETC

RSC-364DWF

Speech Recognition Microcontroller
ETC

RSC-364QFP

Speech Recognition Microcontroller
ETC

RSC-4128

Speech Recognition Processor
ETC

RSC-4128100LQFP

Speech Recognition Processor
ETC

RSC-412864LQFP

Speech Recognition Processor
ETC

RSC-4128DIE

Speech Recognition Processor
ETC