msm6679al-110 Oki Semiconductor, msm6679al-110 Datasheet - Page 21

• Manufacturer Part Number: msm6679al-110
• Description: Si/sd Voice Recognizer, Recorder/player, And Speech
• Manufacturer: Oki Semiconductor

¡ Semiconductor
MSM6679AL-110 Voice Recognition Processor
Name Tag Recording
To facilitate SD recognition, the MSM6679AL-110 supports recording and playback of name tags.
Name tags are used to confirm correct responses in SD recognition. For example, in a phone
dialer application, the user associates a “name” (which is recorded into memory) with a phone
number. The MSM6679AL-110 then plays back the name tag so that the user can verify that the
recognized phrase is the correct one.
The VRP stores names tags in memory using an ADPCM compression algorithm with 28 kbps
of speech. The length of a name tag is controlled with a command from the users host MCU
program. The maximum number of name tags possible is 61, but the actual number is dependent
upon record time and memory available. See the section on memory interface for more detail.
Audio Input Interface
A critical item for high-accuracy speech recognition is correct design of the audio input circuit.
A circuit with appropriate gain and frequency responses must be placed between the microphone
and MSM6679AL-110’s A/D input. Oki recommends input gain and a band pass filter with the
following characteristics:
• Four pole Chebyshev high-pass filter, 3 dB point at 225 Hz
• Dual-pole low-pass filter, 3 dB point at 4250 Hz
• Midband gain of 46 dB at 1000 Hz
The above gain and filter characteristics are obtained by using a rail-to-rail quad CMOS op-amp
and one-half supply rail splitter to bias the input signal at 1/2 V
nominal.
DD
The MSM6679AL-110 uses multiple analog inputs to improve sampling quality. An on-chip
analogy to digital (A/D) conversion unit transforms the analog signal to a digital data stream.
Audio Output Interface
The MSM6679AL-110 also provides the VOICEOUT1 PWM output. The MSM6679AL-110 uses
ADPCM to generate voice or sound-effect output. ADPCM represents an improvement over
conventional PCM techniques in that it adaptively changes the quantizer step (scale factor) to suit
the waveform being encoded. The result is more efficient memory usage with no loss of quality.
Careful selection of the components for internal and external output filters and amplifiers is
recommended. An incorrect choice would impair the original quality. This consideration equally
includes:
• Careful separation of analog and digital lines
• Grounding of analog lines at both ends
• Further adequate separation from high-speed digital circuits to avoid distortions thereof
Memory Interface
The memory control section manages RAM and/or ROM devices in two 64-Kbyte memory
spaces, in conjunction with internal memory for voice templates and working memory. Some
versions work with no external memory, some have some external RAM, some use only external
EPROM, and some use external memory in conjunction with both internal ROM and RAM. The
MSM6679AL-110 requires a minimum of 32 Kbytes SRAM and 16 Kbytes ROM.
The following table shows vocabulary sizes and playback facilities for various configurations.
19