msm6679al-110 Oki Semiconductor, msm6679al-110 Datasheet

no-image

msm6679al-110

Manufacturer Part Number
msm6679al-110
Description
Si/sd Voice Recognizer, Recorder/player, And Speech
Manufacturer
Oki Semiconductor
Datasheet
MSM6679AL-110
Voice Recognition Processor
FIRST EDITION
ISSUE DATE: Nov. 1998

Related parts for msm6679al-110

msm6679al-110 Summary of contents

Page 1

... MSM6679AL-110 Voice Recognition Processor FIRST EDITION ISSUE DATE: Nov. 1998 ...

Page 2

... System Configuration Example ....................................................................................................... 14 Functional Description ...................................................................................................................... 15 Voice Recognition ............................................................................................................. 15 SI Recognition ................................................................................................... 15 SD Recognition ................................................................................................. 18 Name Tag Recording ........................................................................................................ 19 Audio Input Interface ....................................................................................................... 19 Audio Output Interface .................................................................................................... 19 Memory Interface .............................................................................................................. 19 External Voice Synthesis Control ................................................................................... 21 Serial Interface ................................................................................................................... 22 MSM6679AL-110 Slave-Mode API .................................................................................................. 23 Command Summary ........................................................................................................ 24 Command Descriptions ................................................................................................... 28 Asynchronous Serial Protocol Example ........................................................................ 40 ¡ Semiconductor ...

Page 3

... Oki-defined test procedure). For SD recognition, the MSM6679AL-110 stores SD vocabulary templates, as defined by the user, in external SRAM. The MSM6679AL-110 can create SD vocabularies words each, with each word using approximately 50 bytes. In addition to providing voice recognition capabilities, the MSM6679AL-110 integrates a solid- state recorder/player, speech synthesis functions, and a tone generator ...

Page 4

... Recognition and Input Synthesis Engine Vocabulary Memory Figure 1. MSM6679AL-110 Block Diagram A/D Interface Serial-Mode MSM665x Interface PWM Output Serial Interface IC Reset and Oscillator Inputs Figure 2. MSM6679AL-110 Logic Symbol 2 System Controller Algorithm Memory External Memory Control ADC0 ~ ADC7 VREF A0 ~ A15 NAR BUSY ...

Page 5

... SD 7 N/C 8 N/C 9 VOICEOUT1 10 11 N/C NAR GND 15 N/C N N/C 18 N/C N/C 19 PDC 20 N N/C N N/C Figure 3. MSM6679AL-110 100-Pin TQFP Pinout MSM6679AL-110 Voice Recognition Processor 75 A15 74 A14 73 A13 72 A12 71 A11 70 A10 GND ...

Page 6

... MSM6679AL-110 Voice Recognition Processor MSM6679AL-110 Alphabetic Pin List Name # Name # A0 60 A10 A11 A12 A13 A14 A15 ADC0 ADC1 ADC2 ADC3 85 4 Name # Name # Name ADC4 RAMPAGE0 D5 56 RAMPAGE1 ADC5 87 RDRAM ...

Page 7

... Semiconductor Figure 4. MSM6679AL-110 100-Pin Package Mechanical Drawing MSM6679AL-110 Voice Recognition Processor 5 ...

Page 8

... NC (do not connect) Reserved. This pin is reserved for future use and must be left open. 12 NAR Input MSM665x Next Address Request. This pin signals to the MSM6679AL-110 that the external speech synthesis engine is ready for another command Digital Power Power GND Digital Ground Ground ...

Page 9

... When used with an external clock, the external clock is applied to this input. Oscillator 1. When the MSM6679AL-110 uses a crystal oscillator, this output is the oscillator output pin. The pin is then connected to one side of a crystal and load capacitor. When used with an external clock, this output is left unconnected ...

Page 10

... MSM6679AL-110 Voice Recognition Processor Pin # Pin Name Signal Type 47 NC (do not connect) Reserved. These pins are reserved for future use and must be left open. 48 RDRAM 49 Output RAM Read. This is a strobe signal for direct connection to an external RAM's RD input. When asserted LOW, this signal indicates that the MSM6679AL- 110 is ready to read data from RAM ...

Page 11

... V Digital Power Power VREF Analog Power Analog Power. The MSM6679AL-110's on-chip A/D converter uses this Reference Voltage analog power when converting an analog signal into digital samples. Also this is used as an analog reference voltage. 82 ADC0 Analog Input Analog Input. These eight inputs are tied together and serve as the analog input ...

Page 12

... MSM6679AL-110 Voice Recognition Processor ELECTRICAL SPECIFICATIONS Absolute Maximum Ratings Parameter Digital power supply voltage Input voltage Output voltage Analog power/reference voltage Analog input voltage Power dissipation Storage temperature 1. Permanent device damage may occur if ABSOLUTE MAXIMUM RATINGS are exceeded. Functional operation should be restricted to the conditions as detailed elsewhere in this data sheet ...

Page 13

... Output capacitance C O Analog reference power I REF supply voltage Power consumption Typical condition 25˚C. MSM6679AL-110 Voice Recognition Processor Condition Min Applied to D0-D7 0.44 V Applied to all other I/O 0.80 V Applied to D0-D7 –0.3 Applied to all other I/O –0.3 Output current = –400 mA, applied to D0-D7, WRRAM, RDRAM and ...

Page 14

... MSM6679AL-110 Voice Recognition Processor AC Characteristics External Data Memory Control (VDD = 2 -30 ~ 70˚C) Parameter Cycle time Clock pulse width (HIGH level) Clock pulse width (LOW level) RDRAM pulse width WRRAM pulse width RDRAM pulse delay time WRRAM pulse delay time Address set-up time ...

Page 15

... Semiconductor Timing Diagram t CLK t fWH RDRAM A0 - A15 WRRAM A0 - A15 CLK WRRAM RDRAM A0 - A15 RAP0 - 15 DIN0 - 7 DOUT0 - 7 Figure 5. RAM Read/Write Timing MSM6679AL-110 Voice Recognition Processor CYC t fWL RAP0 - DIN0 - ACC RAP0 - 15 ...

Page 16

... MSM6679AL-110 Voice Recognition Processor SYSTEM CONFIGURATION EXAMPLE SRAM Figure 6. MSM6679AL-110 System Configuration Example 14 FLASH MSM6679AL-110 MSM66P54 ¡ Semiconductor ...

Page 17

... The vocabulary phrase that achieves the highest score (or lowest distance) is judged to match the input phrase, assuming that the score exceeds a predetermined threshold. 5. Via a special command, the MSM6679AL-110 can also return the scores of the input against all defined vocabulary phrases for recognition. This feature allows external host software to select the next best match, if the closest match is not contextually logical ...

Page 18

... MSM6679AL-110 Voice Recognition Processor MSM6679AL-110 Control Vocabulary Sub-Vocabulary 1 Sub-Vocabulary 2 Phrase Index Phrase Index A/C 1 Low Fan 2 Medium Temperature 3 High Timer 4 Increase Service 5 Decresse Help 6 Set Select 7 Reset — — Cancel — — Clear — — Recall — — On — — Help MSM6679AL-110 Direction Vocabulary ...

Page 19

... Sechs Sieben Acht SI vocabulary generation starts with collecting reference utterances from 400 speakers with: • An equal mixture of males and females • Accents from all regions of the country of intended use • ~15% non-native speakers. MSM6679AL-110 Voice Recognition Processor Index Phrase Index Phrase 1 Hyoujun ...

Page 20

... REJ SD Recognition In SD recognition mode, the MSM6679AL-110 can be trained to recognize words. The MSM6679AL-110 can support multiple speakers by switching vocabularies, but only one speaker’s vocabulary should be active at one time. The end user enrolls a phrase in the MSM6679AL-110’s vocabulary by recording the phrase three times or more ...

Page 21

... Name tags are used to confirm correct responses in SD recognition. For example 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. ...

Page 22

... SD recognition vocabularies are volatile in these configurations. 3. Per download. Vocabulary swapping by host permits unlimited vocabulary size. The MSM6679AL-110 supports 32 Kbytes of RAM, and Kbytes of ROM (EPROM or Flash) per bank in separate memory spaces. For accessing the ROM and RAM address spaces, the MSM6679AL-110 provides the separate Write RAM (WRRAM) and Read RAM (RDRAM) signals ...

Page 23

... Figure 7. MSM6679AL-110 External Memory Map External Voice Synthesis Control The MSM6679AL-110 is capable of interfacing to the MSM665x family of Oki ROM, OTP, or external EPROM speech synthesizers, allowing for up to 260 seconds of high-quality voice and sound effects. The following table indicates the speech capabilities of the MSM665x family. ...

Page 24

... BUSY - Asserted LOW during MSM665x device playback. The MSM6679AL-110 F50Bh and F10100xxh commands select this signal for MSM665x command polling. • NAR - Next Address Request status signal. By default, the MSM6679AL-110 uses this signal to poll commands to the MSM665x. The F51Bh, F480h, and F440h commands select NAR for polling. • ...

Page 25

... Messages are of variable length. All messages consist of an even number of bytes. Opcodes consist of exactly four bytes, with values between F000h and FEFEh. Operand bytes may take values from 0000h to FFFFh. The MSM6679AL-110 issues a return code for many of the host commands. The return code generally consists of the same opcode, followed by data indicating success of failure of the operation ...

Page 26

... Set output volume (03h = minimum, FEh = maximum). F500 Status request. F501 Select last FLASH bank for SI recognition. Request F510 Select download RAM bank for speaker independent/signal processing (SI/SP) template area. F520 Set MSM6679AL-110 power down mode. 24 ¡ Semiconductor Description Default (Hex) — 8000 4A00 F100 F123 0101, 0202 ...

Page 27

... F210h Acknowledgment F208h F204h F202h F201h Speech Ack F400h MSM6679AL-110 Voice Recognition Processor Description Description Record time = tm*14 msec. High and low bytes of SP/SI origin address. High and low bytes of SD origin address. High and low bytes of triggering origin address. SP table Xh selected. Invalid message received. ...

Page 28

... Rejection: bad signal/noise ratio. Rejection: reason uncertain. Aborting SD Listen mode. After SD utterance search: not found. Rejection. Sort completed. After SD utterance search: empty. Rejection: MSM6679AL-110 SD memory full/empty. After SD utterance search: in use. Utt = Utterance ID triggered. Utterance ID, high/low byte of distance. Utterance ID, high/low byte of distance to utterance 1... ...

Page 29

... If in response to an N-byte download request, the MSM6679AL-110 then receives N bytes ( even odd) of data from the host odd and N+1 bytes are received, only N bytes are written to MSM6679AL-110 memory response to an upload, the MSM6679AL-110 then sends N bytes ( even odd) of data to the host. 5. ...

Page 30

... XXYYh = high (XXh) and low (YYh) bytes of requested address. The MSM6679AL-110 uses and returns an even F104h XXYYh address outside the MSM6679AL-110 work space that is as near as possible to the requested address. Leave this parameter at its default value unless you are using an Oki custom SI vocabulary and are instructed to alter triggering origin ...

Page 31

... F12Yh Set SD Origin command (F103h). After setting the table number and returning the resultant value, the MSM6679AL-110 checks the validity of the SP header. If the header is invalid, an error message is returned. Set this value to (NSI +1), where NSI is the number of SI subvocabularies. ...

Page 32

... Power-On/ Values Reset Value After power-on, the MSM6679AL-110's mode corresponds to that after issuing a F20C command. This mode may NOT be the optimum condition for most situations, so the user is advised to carefully understand the desired condition and develop a suitable command for the application at hand. ...

Page 33

... These beeps do not cause a F400h message to be issued to the host MCU. When set to 1, the MSM6679AL-110 beep can help a user avoid speaking before the MSM6679AL-110 is ready. This mode is normally used with a digits vocabulary to pace the user and confirm each utterance reception ...

Page 34

... MSM6679AL-110 Voice Recognition Processor Recognize Opcode Action Stop Listening. This command causes the F300h MSM6679AL-110 to exit Listen mode, whichever was active. Start SI Listen Mode. For all the following opcodes, the MSM6679AL-110 per-forms SI recognition on incoming utterances, using SI vocabulary Y. The vocabulary Y is identified by one of 15 sets, thus ...

Page 35

... MSM6679AL-110 to play back a name tag from external memory F401h ~ sound is defined for a selected index, the F43Dh MSM6679AL-110 plays a beep. See the Record commands for information on creating name tags. Speak Phrase from Low Internal Memory sound is defined for a selected index, the MSM6679AL-110 plays a beep ...

Page 36

... Speak (Continued) Opcode Action Speak Phrase from High Internal/External Memory sound is defined for a selected index, the MSM6679AL-110 plays a beep. The default phras-es supplied with the MSM6679AL-110 in the larger upper playback memory area are listed below. "0" simulated DTMF tone. ...

Page 37

... Set maximum output level. Request Opcode Action Status Request. This command causes the F500h MSM6679AL-110 to return a 2-byte value indicating its current status. F501h Select last FLASH bank for SI recognition. Select download RAM bank for SI/SP template area. This command enables the download ...

Page 38

... Immediately after receiving parameter NL, the MSM6679AL-110 responds with a message to indicate acceptance or denial of the transfer request. Acceptance is indicated by F5A0h. Denial is indicated by a F8XYh. At the end of an accepted transfer, the MSM6679AL-110 re-sponds with a message to confirm or deny valid completion of the transfer. Valid completion is indicated by F5C0h. F880h Invalid message received ...

Page 39

... Request (Continued) Opcode Action Initialize in Background. Background noise initialization is performed for 500 ms. The MSM6679AL-110 requires this command prior to recognition for noise vector subtraction during the utterance sampling period. Use the background initialization command whenever there is a change in the background noise level. ...

Page 40

... MSM6679AL-110 Voice Recognition Processor Record Opcode Action Set Name Tag Length, Set MSM665x Busy Mode ON. Name tag record length is set by XXh, with XXh defining record length in 14-ms F101h intervals. 00XXh The maximum record length of FFh yields a recording interval of 3.57 sec. The default value is 1.2 sec. ...

Page 41

... The SD F9XYh vocabulary memory is searched for utt. no. XYh not found and if sufficient SD memory exists, the MSM6679AL-110 prepares to add utterance number XYh to the vo-cabulary. Enroll SD Utterance. This command starts MSM6679AL-110 SD Listen mode, then uses the next captured utterance to start or update ...

Page 42

... Asynchronous Serial Protocol Example All messages to the MSM6679AL-110 (except downloads and uploads) are echoed, but replies from the MSM6679AL-110 to the host are not echoed by the host. This arrangement facilitates manual communication with the MSM6679AL-110 using standard terminals. The following table illustrates the range of MSM6679AL-110 functions. ...

Page 43

... MSM6679AL-110 accepts request, signals in progress. MSM6679AL-110 sends 46 bytes. MSM6679AL-110 indicates upload complete. Set up MSM6679AL-110 Host requests set SP table 3. for SI recognition. MSM6679AL-110 selects SP table 3 and confirms. Host initializes MSM6679AL-110. MSM6679AL-110 acknowledges. SI recognition. Host starts SI recognition, vocabulary 1. MSM6679AL-110 recognizes utterance 3. Host starts SI recognition, vocabulary 2. ...

Page 44

... Host prepares MSM6679AL-110 to train SD utterance 2 Memory is empty and ready to train. Pass 1; host sends SD enroll command. SD utterance 2 initialized. Pass 2; host sends SD enroll command. MSM6679AL-110 updates SD utterance 2. Pass 3; host sends SD enroll command. MSM6679AL-110 signals operation completed. SI recognition of control Host starts SI recognition, vocabulary 1. word. ...

Page 45

... Name tag playback. Host sets volume to max. level. Host commands play back name tag 1. MSM6679AL-110 signals playback OK. Sound playback. Host sets output volume to mid point. Play MSM6679AL-110 internal sound 1. Play back sound from MSM6654. MSM6679AL-110 Voice Recognition Processor Action Voice Input "Jane Doe" ...

Page 46

... E2Y0001-28-41 MSM6679AL-110 Voice Recognition Processor NOTICE 1. The information contained herein can change without notice owing to product and/or technical improvements. Before using the product, please make sure that the information being referred to is up-to-date. 2. The outline of action and examples for application circuits described herein have been chosen as an explanation for the standard action and performance of the product ...

Related keywords