CWDSP56800 Freescale Semiconductor, CWDSP56800 Datasheet - Page 3

• Manufacturer Part Number: CWDSP56800
• Description: Tools, Development Kit
• Manufacturer: Freescale Semiconductor

Specifications of CWDSP56800

• Supported Families: DSP56F805
• Core Architecture: 56800/E
• Mcu Supported Families: DSP56800
• Rohs Compliant: NA
• Lead Free Status / RoHS Status: na

Design Challenges
The digital processing capabilities of a
hybrid MCU allow voice control to
penetrate to embedded systems.
Practically any new product containing a
hybrid MCU can be controlled by voice.
A system for control of lights and heating
(see the block diagram on page 1) can
be generalized for controlling any device
with a voice command set. Algorithms
are based on hidden Markov models
(HMM) adapted to the hybrid MCU.
Freescale Semiconductor Solution
Speech Signals in Real Time
The hardware for voice control can be
produced right now because the
DSP56F805 device has 32.5K Flash
program memory, allowing the
recognition algorithm to be upgraded
easily.
The pivotal feature for speech
processing is the ability to parameterize
speech signals in real time. This is
provided in the SDK library.
Voice Command Set Capability
The voice command set must be
designed with respect to speaker
dependence. Assuming the proposed
system will be used by two people, the
SG2041-3
ADC Interrupt
1. Speech Sample -> Buffer
2. Sensor Sample -> Buffer
SOFTWARE MODULE FUNCTIONS
GPIO Interrupt
1. Read Key
2. Read Sensor
Timer Interrupt
Realtime Clock
Input
command set could be the following
words: light, dark, heat, cold (recorded
four times individually for each speaker),
0 through 9, time, and temperature
(recorded by one authorized person).
The speaker dependence characteristic
is advantageous when controlling
devices.
3 Ways to Access Functionality
All proposed system-controlling
functions are accessible by voice
command, manual switch, or by keypad
input (in a noisy environment). Each
microphone has a corresponding switch
or button and a corresponding lamp unit.
The arbitration process, or choosing the
controlled device, is solved by software.
The controller continuously samples all
analog-to-digital converter (ADC) inputs,
the affiliated software process
determines if there is a speech on input,
and the processor assigns an actual
device. The ADC on the DSP56F805
has two modules, each multiplexed to
four pins; a total of eight sources of
analog signal. Optionally, one ADC pin
can be connected to a phone through a
subscriber line interface circuit (SLIC);
one pin is connected to a temperature
Speech Processing
1. Parameterization
2. HMM
3. Verification/
Sensor State Evaluation
1. Arbitration
2. Start Action (Light,
Identification
Siren, Phone Call)
Core Algorithms
SPI Interrupt
Sample from Buffer ->
DAC
Timer Interrupt
Light/Siren Off
sensor. The remaining six pins can be
connected to microphones. The ADC
resolution is 12-bit, and maximum
sampling frequency is 800 kHz, time-
multiplexed sampling of all eight ADC
channels. To minimize memory
requirements, a sampling frequency of
8 kHz is recommended for all speech
channels.
Optional Phone Control of Heating
As an option, heating can be controlled
by phone. The recognition process is the
same. If the speech reference is
recognized, the processor returns it as
an audio signal to confirm the validity of
the recognition. The controller is
connected to a phone through a SLIC.
Internal ADC is used at the input side
and an external digital-to-analog
converter (DAC) is used at the output
side. The DAC and the DSP are
connected through a serial peripheral
interface (SPI).
Software Module Functions
Module functions are provided by
software and should be programmed in
C language. Proposed layering of the
software and interrupt usage are shown
in the figure below.
Output