User`s Manual
Contents
1. Core Features Programmable 5 channel synthesizer Natural phonetic speech synthesis DTMF and other sound effects Programmable control of pitch rate bend and volume Programmable power up or reset announcements Multiple modes of operation Simple interface to microcontrollers Simple Stand Alone operation Three programmable digital outputs Internal 64 Byte buffer Internal programmable EEPROM Extremely low power consumption Low pin count Multiple case styles available SS S7 Special Features KIA e Three multipurpose programmable digital outputs allow the SpeakJet to control external devices based on timing of the sound output Control of devices may include lights motors or even launch model rocket after a count down sequence gt An internal clock oscillator provides for a truly Self Z Contained sound system Simply connect the SpeakJet e to a power supply and a speaker to hear it speak An internal user programmable EEPROM allows for programming of up to 16 complex phrases or sound sequences These may be played back once or looped many times in response to events Phrases may call other phrases sounds or controls with nesting up to 3 levels deep No special equipment is required to program the internal EEPROM only a serial connection is required SEP x Yi Interface Options 3 WY e CPU Control Single Wire Serial Input from microprocessors such as the OOPic2. Nat v ZO speak e ural Speech amp Complex Sound Synthesizer User s Manual February 26 2004 Z lt Voice PWM Out 0 Ready Out 1 Speaking Out 2 Buffer Half Full V MO M1 Reset Rcx General Description The SpeakJet is a completely self contained single chip voice and complex sound synthesizer It uses Mathematical Sound Architecture MSA technology to control an internal five channel sound synthesizer to generate on the fly unlimited vocabulary speech synthesis and complex sounds The SpeakJet is preconfigured with 72 speech elements allophones 43 sound effects and 12 DTMF Touch Tones Through the selection of these MSA components and in combination with the control of the pitch rate bend and volume parameters the user has the ability to produce unlimited phrases and sound effects with thousands of variations at any time Z The SpeakJet can be controlled simultaneously by logic changes on any one of its eight Event Input lines and or by a single I O line from a CPU such as the OOPic Basic Stamp or PC allowing for both CPU Controlled and Stand Alone operations gt Other features include an internal 64 byte input buffer Internal Programmable EEPROM three programmable Outputs and direct user access to the internal five channel sound synthesizer lt More information can be found at www SpeakJet com v 1 ME Lr DIR Lo
3. A simple two pole low pass filter is all that is required before an amplifier to obtain high quality sound output See Figure 1 lt AM QN Mode Select When a High logic level is placed on Mode Select MO and RESET is forced the SpeakJet starts a Demo mode where all internal allophones and sound effects are played in a loop This loop plays if the buffer is empty and will suspend until the buffer is empty again if data is sent to the buffer When MO is brought back to a low state the demo mode is suspended If MO is brought back to a high state the demo mode continues from the point where it was suspended A RESET with a low logic level on Mode select MO will discontinue the Demo Test mode and place the SpeakJet back to normal operation NIS If the SpeakJet is in the Demo Test mode and a momentary Low logic level is placed on Mode Select M1 the Speakjet starts a Sonar Ping sound to indicate the Baud Rate Configuration mode is selected At this time the Speakjet is waiting for a serial sync character hex 55 to set the internal clock for the baud rate that is to be used Once the baud rate is captured the Speakjet automatically stores value this value in the EEPROM for future reference even after a power down condition and returns to the demo mode A Normal operation requires that MO to be Low logic level and M1 to be a High logic level NY amp Serial Data lt iS 50k Pot 27k B ES
4. Basic Stamp or any other computer system equipped with a serial port Stand Alone Eight Event Inputs for execution of up to sixteen phrases sound effects or control functions without a microcontroller RC Input Two Servo Inputs for execution of up to four phrases sound effects or control functions via wireless model Airplane or Car Radio Control receivers ez SpeakJet Pin Details Oz 0 ON va of Pin Description Functional Details 1 RC1 D7 RC1 Input Event Input 7 Electrical Specifications Supply voltage 2 0 to 5 5 VDC Supply Current Idle lt 5ma Plus loads Speaking lt 5ma Plus loads Sink Source Current Outputs 25ma Inputs levels High Supply Low GND EEPROM Max write cycles Typical 1 000 000 times Mechanical Specification 2 RCO D6 RCO Input Event Input 6 3 D5 Event Input 5 4 D4 Event Input 4 5 GND Ground 6 D3 Event Input 3 7 D2 Event Input 2 8 D1 Event Input 1 9 DO Event Input 0 10 RCX TTL Serial Input Active High 11 RESET Master Reset Active Low 12 M1 Baud Select 13 MO Demo Mode Active High 14 V Power input 2 0 to 5 5 volts DC Thermal storage 60 to 140 Degrees C 15 OUT2 Buffer Half Full Output 2 External Buffer Half Full Internal Thermal operating 50 to 120 Degrees C 16 OUT1 Speaking Output 1 External Currently Speaking Internal 17 OUTO Ready Output O External Spe
5. English language and give suggestions for the appropriate times that they are used For example there are 4 B sounds BE and BO sound good in initial position as in the words Beep and Box while EB and OB sound good in final position as in Rib and Club A vowel modifier is also associated with the BE BO EB and OB sounds The BE is used when followed by a front vowel sound and the BO is used when followed by a back vowel sound Note that either can be used when followed by a central vowel sound Also note that a B sound in the middle of a word can be either a initial position or a final position sound For example the word rabbit is it RAB IT or RA BIT Local dialect can be the deciding factor in cases like this Typically the first allophone in a phrase is longer than the rest of the allophones used with the phrase Therefore to create an initial SS you can use the SLOW function before the SS allophone which will cause the SS sound to play for 1 and 1 2 the normal speed If an even longer sound is needed then the SS allophone can be played twice Note that this cannot be done with all allophones Repeating Diphthongs for example will cause the diphthong slide to play twice which will not result in just a longer sound Diphthongs are perceived as single vowel sounds but in reality they are sliding sounds that transition from one vowel sound into another vowel sound For example the A sound
6. Fall Comb Over Hold Book Could Should Food June Milk Famous Nip Danger Thin No Snow On Think Ping Hung Song Lake Alarm Lapel Clock Plus Hello Wool Sweat Ray Brain Over Clear Hear Year Hair Stair Repair Fir Bird Burn Part Farm Yarn Corn Four Your Gate Ate Ray Mice Fight White Boy Toy Voice Sky Five Yes Yarn Million Saddle Angle Spell Cute Few Brown Clown Thousand Two New Zoo Our Ouch Owl Go Hello Snow Dodge Jet Savage Vest Even Zoo Zap Azure Treasure There That This Bear Bird Beed Bone Book Brown Cab Crib Web Bob Sub Tub Deep Date Divide Do Dust Dog Could Bird Bud Food Get Gate Guest Got Glue Goo Peg Wig Dog Peg Church Feature March Help Hand Hair Hoe Hot Hug Who Whale White Food Effort Off See Vest Plus Zz abies Lo 70 200 200 190 200 185 165 200 225 185 170 140 180 170 170 200 131 70 70 70 70 70 45 45 10 10 45 45 10 10 55 55 55 55 70 70 70 70 70 40 Silence Silence Silence Silence Silence w Trans Silence w Trans Silence w Trans Voiced Long Vowel Voiced Short Vowe Voiced Long Vowel Voiced Short Vowel Voiced Short Vowel Voiced Short Vowel Voiced Short Vowel Voiced Short Vowel Voiced Short Vowel Voiced Long Vowel Voiced Short Vowel Voiced Long Vowel Voiced Nasal Voiced Nasal Voiced Nasal Voiced Nasal Voiced Nasal Voiced Resonate Voic
7. Typical connections Serial Control 27K A 10 uf Serial TTL Data a Audio Amp Figure 1 SpeakJet Typical connections X Mor E Serial Data is the main method of communicating with the SpeakJet By utilizing an internal Baud rate generator with an Auto Baud Rate Detector a wide range of serial data can be read and interpreted to execute voices sounds or commands The serial data can also be used to program the internal EEPROM The SpeakJet does not accept actual RS 232A signal levels and will be damaged if attempted to input these levels In order to read data from a RS 232A type of serial data stream a level shifter line receiver must be used This can be as simple as a small transistor or more robust device like the MAX232A from Maxim The SpeakJet serial configuration is fixed at Non Inverted No Parity 8 bits and 1 stop bit N 8 1 The Speakjet can accept Baud rates from 300 to 19200 The factory default setting is 9600 baud This baud rate can be changed any time by placing the SpeakJet into the Baud Rate Configure Mode The Speakjet may need to have the Baud Rate Configure Mode executed if it is placed in an area where the temperature has changed greatly from where it was last performed This is due to the internal Oscillator drift with a large temperature change NV A X The Speakjet utilizes SCP Serial Control Protocol developed by and licensed from Savage Innovations Makers o
8. used as a general purpose output line for controlling external devices such as indicator lights piezo audio transducers relays MOSFET transistors etc This line is a logical output with either a High or Low voltage level The function of this 2 FRI NR 3 VOLTS N 2 X AA BATTERIES Z lt 120 OHM SPEAKER OR HEADPHONE MAX 25 ma gt Minimum connections amp output line is stored in the EEPROM and the factory default Demo Test Mode G configurations is Speaking and is Active High v 2 N 4 72 au u S9 NOLTE Serial Flow Control OUTO Ready 9 This output line can be used to indicate when the Speakdet is Ready This indicates that the internal self test has passed and the Speakjet is Ready to accept data through one of its many inputs It may also be used as a general purpose output line for controlling external devices such as indicator lights piezo audio transducers relays MOSFET transistors etc This line is a logical output with either a High or Low voltage level The function of this output line is stored in the EEPROM and the factory default configurations is Ready and is Active High SN Voice PWM SD OX Voice PWM line is an output line with a square wave of 8KHz The duty cycle of this line is varied by the modulation of the sound output This duty cycles varies by as much as approximately 20 to 80 depending on the level of output represented by the audio wave
9. As the SpeakJet executes the commands stored in the input buffer the buffer is freed and as 32 bytes of the buffer becomes available The Buffer Half Full status line will reflect that status This line can be used in a serial connection as the serial port s RTS line A logical Low output indicates that the buffer can accept 32 bytes and a logical High output indicates that the buffer cannot The status lines can also be user programmed to control external devices Each Status line can be individually selected to be and internal Status or user selected logic state ALLOPHONE SPEECH SYNTHESIS PRIMER Introduction The sounds that are used to form the words of any particular language are called phonemes During the articulation of the words of that language variations of phonemes called allophones are vocalized one after another producing the spoken words and Phrases gt The allophone speech synthesis technique provides the ability to synthesize a reproduction of that language by providing a method of playing allophone like sounds one after another in much the same way that the human mouth does The SpeakJet accomplishes allophone speech synthesis via Mathematical Sound Architecture m MSA The function of MSA is to mathematically model the sounds that the human mouth produces as it moves from one position to the next Seventy Two discrete speech sound elements one for each allophone are stored in the SpeakJet s internal MSA Sound Comp
10. ach command received is buffered into a 64 Byte input buffer and executed in the order that they are received First In First Out FIFO In the case of the SCP these commands are executed immediately as they are received and not stored in the input buffer See the section on SCP for further details As the SpeakJet executes the commands stored in the input buffer each command is removed from the input buffer after the command is executed Both the Serial Input line and the 8 Event input lines provide methods of placing command data in the input buffer Serial Input Line a single input line is used to receive command data at the preset baud rate The Rex line is configured as the serial input line and is expecting a TTL level non inverted signal Event Input Lines In the case that the event line is configured to do so a single EEPROM Call command is stored into the input buffer whenever one of the Event lines logic state is changed or in the case of the RC TTL lines when a PWM pulse changes a predetermined amount Status lines provide information on the current operational status of the SpeakJet Ready is active high when the SpeakJet is on and ready to accept commands Speaking is active high when the SpeakJet is currently generating any sound Buffer Half Full is active high when the input buffer is 1 2 full A Buffer Half Full status indicates when the 64 Byte input buffer has more than 32 bytes in it
11. akJet Ready Internal The thermal specifications are preliminary and 18 Voice PWM 8 KHz PWM Voice Output may change as testing is completed VG WY PS AN Q SpeakJet Pin Functional Explanation lt a DA Event Input 0 7 Event Inputs provides a way to execute phrases sound effects or control functions that have been pre programmed into the EEPROM Each pin can be programmed to execute a phrase sound effect or control function on a Low to High going transition or a High to Low going transition This can be combined to allow two phrases or sound effects to be played with one pin or disabled This feature is controlled by an entry into the EEPROM These are logical inputs that require a High or Low voltage level When not used tie to GND gt Um RCO RC1 Input NIS RC input allows the SpeakJet to connect to model radio control receivers and mimic a servo to execute up to four phrases sound effects or control functions These phrases sound effects or control functions can be pre programmed into the EEPROM These are logical inputs that require a High or Low voltage level Oi ADD RCX Receive Serial Data Input Gives external devices a means to communicate with the internal MS A EEPROM and S Control System by TTL serial data stream This is a logical 7 input that requires a High or Low voltage level e g N OY A 2 2 IR RESET gt OS A low on thi
12. e al Io phone n A predictable phonetic variant of aphoneme For example the aspirated T of Top the unaspirated T of sTop and the TT pronounced T J as a flap of baTTer are allophones of the English phoneme T syl la ble n A unit of spoken language consisting of a single uninterrupted sound formed by a vowel diphthong or syllabic consonant alone or by any of these sounds preceded followed or surrounded by one or more consonants syn the sis n The combining of separate elements or substances to form a coherent whole NIS VRA DA FRI NR 64 Byte Synthesizer Buffer Preprocessor User ES M S A Digital l EEPROM Outputs MS A Sound SpeakJet Serial Event Input Port IRC Component Control Database ae SS N E KIA KAA K pb Figure 2 SpeakJet Block Diagram Oy YH D Dy L fy A x WES DA BE Figure 3 SpeakJet 5 Channel Synthesizer Block Diagram KZ Ln Z Dp 2 gt Vip Fa 7 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 DL Phoneme Sound Effect Code List Code Phoneme Samle Words See Even Feed Sit Fix Pin Hair Gate Beige Met Check Red Hat Fast Fan Cotten Luck Up Uncle Hot Clock Fox Father
13. ecifies the desired pitch The vocalization pitch is what makes a voice sound High pitched or Low pitched For singing the pitch has a range of 3 full octaves 32Hz to 240hz The Vocalization Pitch works only on sounds that are voiced Pitches can range from 0 to 255 The default is 88 Note that anything under 30 starts to sound like clicks instead of a voice Also Note that a value of 0 0 Hz and thusly will not actually vocalize 23 Bend X This command sets the frequency Bend A value will need to be sent after the Bend command that specifies the desired Bend The frequency Bend adjusts the output frequencies of the oscillators This will change the voicing from a deep hollow sounding voice to a High metallic sounding voice Bends can range from 0 to 15 The default is 5 24 PortCtr X This command sets the Port Control Value A value will need to be sent after the PortCtr command that specifies the desired function of the output lines The Output line control bits are binaurally encoded where a 1 indicates that the output function is chip controlled and a 0 indicates that the output function is user controlled Bit O corresponds to OUTO etc PortCtr values can range from 0 to 7 The default is 7 25 Port X This command sets the Port Output Value A value will need to be sent after the Port command that specifies the desired state of the output lines When the Output line control bits are set to 0 the corresponding port b
14. ed Resonate Voiced Resonate Voiced Resonate Voiced R Color Vowel Voiced R Color Vowel Voiced R Color Vowel Voiced R Color Vowel Voiced R Color Vowel Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Diphthong Voiced Affricate Voiced Fictive Voiced Fictive Voiced Fictive Voiced Fictive Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiced Stop Voiceless Affricate Voiceless Fricative Voiceless Fricative Voiceless Fricative Voiceless Fricative Voiceless Fricative SI A ES WS A So Sweat Ship Fiction Leash Thin month Part Little Sit To Talk Ten Parts Costs Robots Can t Clown Key Comb Quick Fox Speak Task Book Took October People Computer Pow Copy OONDAARWNM O Sonar Ping Pistol Shot WOW A MS 40 50 40 50 70 170 55 55 55 45 99 99 80 80 80 80 80 80 80 80 80 80 300 101 102 540 530 500 135 600 300 250 200 270 280 260 300 100 104 100 270 262 160 300 182 120 175 350 160 260 95 75 95 95 95 95 95 95 95 95 95 95 95 95 CAA Table A Voiceless Fricative Voiceless Fricative Voiceless Fricative Voiceless Stop Voiceless Stop Voiceless Stop Voiceless Stop Voiceless Stop Voicel
15. ess Stop Voiceless Stop Voiceless Stop Voiceless Stop Robo Robo Robo Robo Robo Robo Robo Robo Robo Robo arm arm arm arm arm arm arm arm arm arm Beeps Beeps Beeps Beeps Beeps Beeps Beeps Beeps Beeps Beeps Biological Biological Biological Biological Biological Biological Biological Biological Biological Biological DTMF DTMF DTMF DTMF DTMF DTMF DTMF DTMF DTMF DTMF DTMF DTMF Miscellaneous Miscellaneous Miscellaneous Silence PPPPP gt P gt gt gt gt SY Control Codes Character ASCII Use SpeakJet Use Null char Start of Header Start of Text End of Text End of Transmission Enquiry Acknowledgment Bell Play Next Sound Fast Backspace Play Next Sound Slow Horizontal Tab Line Feed 3 Vertical Tab Form Feed Carriage Return Shift Out Play Next Sound High Shift In Play Next Sound Low Data Link Escape Wait XON Device Control 1 Device Control 2 Play Next Sound Soft XOFF Device Control 3 Device Control 4 Volume X Negative Acknowledgement Speed X Synchronous Idle Pitch X End of Trans Block Bend X Cancel PortCtr X End of Medium Port X Substitute Repeat X Escape 7 File Separator Call Phrase X Group Separator Goto Phrase X Request to Send Delay X Unit Separator Reset Defaults Silence No Sound End of Phrase Control Codes Details 0 6 Pauses Pauses of various du
16. f the OOPic to give the end user access to all of the low level registers for the Control EEPROM Synthesizers and the digital Outputs Y NIIT LORA SRY General Operation The SpeakJet is an advanced CMOS LSI device that incorporates six basic internal modules Figure 2 5 Channel Synthesizer For generating both voice and sound effects Sound Component Database and Mathematical Sound Architecture MSA engine that controls the 5 Channel Synthesizer Chip control I O which consists of 1 Serial input line 8 Event input lines 3 output lines and 2 mode select input lines 64 Byte input buffer User EEPROM A Serial Control Protocol SCP supervises the incoming serial commands When powered on the SpeakJet will first read the Mode Select line MO to determine if the SpeakJet is to operate in Demonstration Mode Once the Initial Mode is set note that this only occurs at power up the input buffer is cleared the status lines are set and the SpeakJet begins operating and the ready line is brought high During operation the SpeakJet waits for incoming command data which specifies which sounds the SpeakJet is to produce and how to produce them The commands sent to the SpeakJet can be any of 256 commands There are 7 operational groups of commands SCP Allophones Sound Effects DTMF Pauses Levels and Controls Each operational group will behave in different ways With the exception for SCP commands e
17. igible sounding speech with the SpeakJet e First there is no one to one correspondence between written letters and speech sounds Each sound in a language may be represented by more than one letter and conversely each letter may represent more than one sound See the examples in Table A Because of these spelling irregularities it is necessary to think in terms of sounds not letters when creating phrases QN Secondly speech sounds are acoustically different depending upon their position within a word For example the initial B sound in Beep is acoustically different from the B sound in Box The B sounds differ due to the influence of the vowels which follow them Third the human ear may perceive the same acoustic signal differently in the context of different sounds Therefore an allophone may sounds slightly different when used in various phrases Fourth some sounds in words are not actually pronounced and or others are added when followed or preceded by certain other words For example the two words Night and Time Played separately they both will articulate the T sound However played together they need to be articulated as Ni Time OA Fifth The Vocalization Pitch Play Rate and Frequency Bend parameters are just as important as the selection of the phonemes used For instance playing a phase that does not change the pitch at any point in time sounds very monotonic and robotic While increasing the Pi
18. in Make isn t a single vowel sound It is a diphthong that transitions from the EY sound to the IY sound and thusly the name for the diphthong is EYIY This sliding sound is repeated when diphthongs are repeated For example EYIY EYIY does not have the same affect as IY IY Two IY s played back to back will produce an IY sound that lasts twice as long while the EYIY EYIY will just repeat the slide Five of the R sounds are diphthongs that begin with a vowel and end with the R sound For example the AWRR in alarm and the OHRR in score Determining when to use a diphthong and when to use the independent sound again largely relies on person preference and local dialect Some sounds P B T D K G CH and JH require a brief duration of silence before them For these the silence has already been added but you may decide you want to add more To increase the duration of silence the SLOW command can be used and optionally a pause can be inserted before it There are several different types and lengths of pauses that can be inserted in front of an allophone but be aware that different pauses will affect the way the MSA calculates the transitions from the last allophone thought the silence and into the next allophone which may not produce the desired flow of sounds Stress can be accomplished in two ways One is to cause vowels to play for a longer period of time For example in the word extent use the Fast command in fr
19. it is represented on the output line Bit 0 corresponds to OUTO etc Port values can range from 0 to 7 The default is O 26 Repeat X This command sets a number of times to Repeat the next code A value will need to be sent after the Reset command that specifies the number of times to repeat the next command The Repeat range is from 0 to 255 28 Call Phrase X This command specifies which EEPROM phrase to play then to return from This can be nested 3 deep maximum 29 Goto Phrase X This command specifies which EEPROM phrase to play 30 Delay X This command specifies the number of 10ms intervals to delay before continuing on to the next code The Delay range is from 0 to 255 31 Reset This command resets the Volume Speed Pitch and Bend to the default values
20. onent Database Each MSA allophone component was created to duplicate a specific allophone sound made by the human mouth during English speech To produce speech a list of selected allophones is sent to the SpeakJet As the SpeakJet is vocalizing this list of allophones MSA actively and continuously calculates all the sound components of the allophones including the transitional sounds made between the allophones producing the same in between allophone sounds that the human mouth does Any English word or phrase can thusly be created by selecting the appropriate combination of allophones and pauses Further tuning A with the Rate Pitch Bend and Volume parameters adds to the delivery of the phrase and can change the emotion in which the phrase is perceived Stressing the Rate Pitch Bend and Volume parameters to levels outside the human range can result in some interesting sounds that go way beyond what a normal human mouth can produce In addition several other sounds which are included in the MSA Sound Component Database of which some use vocalization modeling and some do not can be integrated into the phrases The result is a system that gives the user the ability to not only produce an unlimited vocabulary but also to produce slang gibberish moans groans yodels and other weird vocalized sounds not normally included in a canned TTS system Producing English Speech Five basic linguistic concepts will help you to create highly intell
21. ont of the EH in the first syllable which is unstressed and a SLOW command or and additional EH in front of the EH in the second syllable which is stressed A second way is to preceded the allophone with the STRESS and RELAX commands The STRESS command duplicates the affect of slightly contracting the muscles of the mouth and the relax command duplicates the affects of slightly relaxing the muscles of the mouth For example STRESS IH sounds more like but not quite the IY sound Likewise RELAX IY sounds more like but not quite an IH sound Note that if you elect to use the STRESS or RELAX command in combination with two EH s then two command will be needed one in front of each of the EH s Remember that you must always think about how a word sounds not how it is spelled For example The N sound in Link is actually the NGE sound and the sounds in the words letter and little use the allophones AXRR and EHEL some sounds may not even be represented in words by any letters as the YUW in computer You will want to experiment with all the possible sounds to discover which version works best in any particular cluster of allophones One last suggestion is that you may want to add a short pause between words and a long pause between clauses pho neme n The smallest phonetic unit in a language that is capable of conveying a distinction in meaning as the M of Mat and the B of Bat in the English languag
22. rations These will cause the volume to ramp down wait a specified amount of time and the ramp back up 1 2 amp 3 ramp the volume while the format frequencies are being changed 4 5 amp 6 wait for silence before changing the format frequencies 0 0ms 1 100ms 2 200ms 3 700ms 4 30ms 5 60ms 6 90ms 7 Fast Plays the next phoneme at 1 2 the time it normally would play 8 Slow Plays the next phoneme at 1 and 1 2 the time it normally would play 14 Stress Plays the next phoneme with a small amount of stress in the voice 15 Relax Plays the next phoneme with a small amount of relaxation in the voice 16 Wait This command will stop the voicing and wait for a start command The Start command can be issued by either sending the SCP start command or by changing the state of one of the input lines that has been previously set to do a Start 18 Soft Plays the next sound with a reduced volume of 50 20 Volume X This command sets the master volume level A value will need to be sent after the volume command that specifies the desired volume Volume levels can range from 0 to 127 The default is 96 21 Speed X This command sets the play speed A value will need to be sent after the speed command that specifies the desired speed Speeds can range from 0 to 127 The default is 114 22 Pitch X This command sets the Vocalization Pitch in Hertz A value will need to be sent after the pitch command that sp
23. s pin forces the internal control circuit to reset and clear the buffer EEPROM stored phrases sound effects control functions and configuration bits are not affected The use of RESET along with the logic states of the Mode Select inputs MO and M1 places the SpeakJet into the different modes of operation This is a logical input that requires a High or Low voltage level oe Be MO M1 Mode select pins places the SpeakJet into one of three modes Normal Operation Demo Test mode and Baud Rate Configure mode These pins are by the internal control system after a power up or a reset is forced These pins are logical inputs that require a High or Low voltage level OUT2 Buffer Half Full This output line can be used for flow control with a serial port by connecting to the DSR line back to the computer or controller It may also be used as a general purpose output line for controlling external devices such as indicator lights piezo audio transducers relays MOSFET transistors etc This line is a logical output with either a High or Low voltage level The function of this output line is stored in the Le EEPROM and the factory default configurations is Buffer lt Half Full and is Active High ez RAD RD O lt OUT1 Speaking This output line can be used to indicate when the Speakdet is Speaking This can be useful to switch on an amplifier or indicate that an audio output is present It may also be
24. tch at the end of a sentence produces a questioning tone Phonemes of English Language Table A contains a chart of all the consonant phonemes of the English language Table A also contains all the vowel phonemes Consonants are produced by creating an occlusion or constriction in the vocal tract which produces an aperiodic sound source If the vocal cords are vibrating at the same time as in the case of the voiced fricatives VV DH ZZ and ZH See Table 5 there are two sound sources one which is aperiodic and one which is periodic gt Vowels are usually produced with a relatively open vocal tract and a periodic sound source provided by the vibrating vocal cords Ce They are classified according to whether the front or back of the tongue is high or low See Table 4 whether they are long or short G and whether the lips are rounded or un rounded In English all rounded vowels are produced in or near the back of the mouth UW UH OW OH OHRR AW How to Use the Allophone Set Recall that aphoneme is acoustically different depending upon its position within a word Each of these positional variants is an allophone of the same phoneme An allophone therefore is the manifestation of a phoneme in true speech signal The following is a brief tutorial on creating speech using the SpeakJet allophone set Note that these are suggestions not rules 4 The allophone set Refer to Table A references the variations of each phonemes of the
Download Pdf Manuals
Related Search