REVIEWS COURTESY OF ZXSR

Cheetah Sweet Talker
Cheetah Marketing Ltd
1983
Sinclair User Issue 20, Nov 1983   page(s) 34

NATURAL SOUNDS BY CHEETAH

Cheetah Marketing has produced a speech output device called the Sweet Talker. It is complete in a black plastic box the same size as the 32K RAM pack for the Spectrum. The unit contains a very powerful amplifier and the tape with it not only auto runs to introduce itself but also shows some very good programming practice.

The instructions are very simple and the unit should not clash with any joysticks or other items, is it uses I/O port address 7. That may clash with some Sinclair peripherals - the Microdrive, for instance - as Sinclair tends to use only one bit of an address going low to operate equipment and using port 7 takes all the bits from B7 to B3 low. Cheetah is looking into that and may modify the address.

Programming is done by using phonetic sounds which allow you to produce the sounds necessary to make up a full word. They are put out as numbers to Sweet Talker in sequence from a DATA statement when you want words spoken. Silences can be included to stop the words or to provide pauses between words - of up to 200ms. The Sweet Talker seems to be about eight to 10 times as slow as others available, which makes it sound more natural. There are ZX-81 and Spectrum versions.

Sweet Talker is available from Cheetah Marketing, 359 The Strand, London WC2 Tel: 01-240 7939. John Menzies and Boots also stock it. The cost is £34.95.


Transcript by Chris Bourne

Sinclair User Issue 29, Aug 1984   page(s) 29

SWEET CHATTER FROM CHEETAH

The Sweet Talker speech synthesiser from Cheetah has been upgraded so that it will now work with Interface 1 and Microdrives.

It uses the popular SP0256 speech chip which contains 64 allophones. An allophone is a sound from which all words are made. They are used by outputting a number to the unit which then 'speaks' the allophone.

In that way they can be strung together to form almost any word in, as the demonstration cassette shows, French, German and English.

Unfortunately while the PCB has been upgraded the demonstration cassette has not. Users who have an Interface 1 attached will find the computer crashes when the program is RUN. To overcome that edit line 5010 to:

OUT 31,a:GOTO 5000.

Despite that, the unit worked very well. A volume control would have been an added advantage but you cannot have everything.

For more details of the Sweet Talker, price £34.95, contact Cheetah Marketing Ltd, 24 Ray Street, London EC1R 3DJ. Tel: 01-278-6954.


Transcript by Chris Bourne

C&VG (Computer & Video Games) Issue 31, May 1984   page(s) 158

HEARING VOICES AGAIN

Latest in the line of speech synthesisers to land on my desk is an offering from Cheetah, which comes as a little black box (don't they all) and sits on the user port at the back of a Spectrum.

Unlike the Commodore Magic Voice cartridge, this one uses a system of allophones. This means that instead of storing whole words, it stores syllables which you can link together to produce absolutely any word in any language.

Although this is harder to program, it is far more versatile and well worth the extra effort. You won't need any software to run the package - it's all on a ROM, so there're no tapes to load.

It does come with a cassette demo, though, complete with sample speech in foreign languages, including German and French. And the Scottish one's pretty funny too.

Sweet Talker, from Cheetah, costs £29.75.


Transcript by Chris Bourne

ZX Computing Issue 14, Aug 1984   page(s) 32,33

TALKBACK: SPEECH UNITS ASSESSED

When you find you get to the stage where you find yourself talkiing to your micro, it helps if it answers back!

From the time the Daleks trundled across our TV screens muttering "Exterminate, Exterminate" in their chillingly inhuman voice, I have been intrigued by speech synthesizers. At the time of writing, and to the best of my knowledge, there are six different units on the market to enable your micro to answer you back. Two others have recently lost their voice, I asked why, but the suppliers aren't talking.

The units on the market at the moment are "Sweet Talker" from Cheetah for both ZX81 and Spectrum, "Micro Speech" from Currah, "Orator" from Fuller, "Vox Box" from Datel Electronics, "Chatterbox" by William Stuart Systems, and "S-Pack" by DCP.

I contacted all of the above companies and those units that were supplied for review I sent to different reviewers for their independent comments.

For your own assessment here's what they thought....

The SWEET TALKER unit (11 x 7.5 x 3cm) plugs easily and firmly into the user port at the rear of the computer and has its own rear card edge connector for a Ram Pack necessary. A demonstration program cassette is supplied and not only gives a good idea of the range of speech sounds possible but also explains how to program the unit and even gives a few ideas for turning games and original programs into 'talkies'. It takes less than a minute to load.

One way to understand how spoken words are formed is to break up the sounds that combine to make a word into allophones, and this unit uses the allophone system to produce words. For instance, the word Cheetah can be broken down to 4 speech sounds or allophones: /CH//IY//TT//AA/. Using the allophone system, each allophone, pause or gap in speech (as at the end of words and sentences) is turned into a number and then given a code. The code is then incorporated within a very simple program using OUT Instruction to 7.

An instruction booklet is supplied with the unit and lists the allophones with everyday words which contain examples of the relevant allophones. Find the allophones required by checking through the word lists, then read across to the column listing the Data Codes and enter them in the first statement, followed by four more short lines and RUN. It talks....

There are several bonuses with this system. Firstly, although experimentation is necessary to achieve maximum clarity, it's great fun and often amusing. The booklet is easy to use. Secondly, it teaches the user a lot about the structure of words, vowels, consonants, word endings and so on. Not unlike the use of phonics in the teaching of reading. In this last respect this system could well find a place in classrooms. One word of warning to parents and teachers; the unit will say anything it's told to - and without so much as a blush!! But when will we be able to program our micros to sing "Daisy, Daisy..."?


REVIEW BY: Colin Christmas

Transcript by Chris Bourne

Your Spectrum Issue 9, Nov 1984   page(s) 52,53,55,57

SPECTRUM SPEAKERS

The word is that attaching any old speech synthesiser to your Spectrum will allow you to have cosy chats together. Henry Budgett determines where this is one of the first signs of madness.

For years science Fiction films and futuristic novels have depicted an era when man and machine can communicate in perfect harmony. The reality, of course, is slightly different. While speech recognition has yet to he fully developed (ACT's latest Rascal notwithstanding), chip-based speech synthesis has been both mastered and available for several years. Until recently the computing power needed to produce human-sounding utterances was substantial. Now almost every home computer is capable of being equipped to talk back to its owner at a price that won't even break the average piggy bank.

SOUNDING OFF

When we speak we produce three distinctly different types of sound. The most obvious are the 'voiced' or vowel-type sounds; oo, ar, ee, and so on. These are produced by air from the lungs making the vocal cords vibrate. The frequency of this vibration determines which vowel sound we hear.

The second group is the unvoiced or 'fricative' sounds; ss, sh, t and ff. Here the air from the lungs rushes past the vocal cords without making them vibrate and the frequency produced is controlled by the positioning of the lips and tongue. Finally there's silence or, to be more precise, the minute gaps that occur within words (for example six, eight) where we change from voiced to unvoiced and vice versa.

FAKING IT

In order to generate speech-like sounds, the electronics designers generally go for one of two methods. The first - and until recently the most common - is synthesis by rule. If the frequencies contained within speech are analysed it's possible to devise a system of rules that allow us to re-create any sound from its basic frequencies.

These 'building blocks' of sound are called phonemes and by using them in various combinations any word can be constructed. The individuality of a human speaker tends to be lost when speech is generated like this but the words can be clearly understood. Because the synthesis rules for each phoneme are built into the equipment, the user has simply to supply a list of phonemes to be spoken. It's then possible to generate complete sentences instantly, simply by calling up a string of stored phoneme commands. In reality these phonemes tend to be called allophones; this is because the various building blocks sound different depending on their positioning within a word or phrase. However the principle's much the same.

The second method for generating speech relies on the fact that the human ear and brain are very good at filling in gaps. The speech we hear over a telephone line is (British Telecom permitting) perfectly understandable. Yet technically the quality - the range of frequencies we can hear - is only one-fifth of what we'd expect from a standard hi-fi system. We understand what's being said only because our brain does the job of filling in the gaps.

With the fall in cost of computer memory it's now possible to convert speech into digital information compressed many hundreds of times by a wonderful mathematical technique called Linear Predictive Coding. The resulting numbers representing the original speech are stored in a ROM. To get any of the stored words out again as speech is easy; we simply give the computer the address in memory of the word and the digital information is recovered and converted back into sound, and because the original speaker's words have been stored, all the personal characteristics remain. That's why Acorn's speech chips for the BBC Micro really do sound like Kenneth Baker.

WHAT'S THE USE?

The commercial uses for speech synthesis are so many and varied that it's just about impossible to list them all. Looking just at the tip of the iceberg it can be used to replace taped announcements at railway stations and airports; in America it's widely used on the telephone system to inform callers of misdialled numbers and engaged or withdrawn services. Speech synthesis units are also being incorporated into cars like Maestros and Montegos as part of the standard instrumentation so, as well as being something of a sales ploy, they can provide warnings the driver can hear without having to take his or her eyes off the road. A major contribution to road safety perhaps?

As far as we are concerned in the home computer and electronic games market, speech synthesis is generally used to enhance games. Scores can be read out and warnings of imminent enemy attack can be given to warn players leaving them free to concentrate on the tactics of the game. Of the five speech units under review here, four of them use the phoneme system and one the stored speech method. Let's take a look at how they succeed in fulfilling their purpose.

SUMMARY

If you're looking for a means of adding a voice to your Spectrum and of incorporating the facility either into games or just for fun, then the Currah MicroSpeech is almost certainly going to be the best buy for you. It's also got the largest number of games already written for it if you prefer to use shop-bought software. Another of its clear advantages over the other units is the addition of a BEEP amplifier for putting the sound through the TV.

For those of you who haven't yet bought a joystick controller or a sound generator and fancy a speech synthesiser at the same time, then the Fuller Box/Orator combination - though expensive - offers the lot in one package.

Serious users of speech output have an equally clear-cut choice. The superior quality offered by the DCP S-Pack's Digitalker chips make this the logical buy for anyone using the Spectrum as an annunciator rather than as a games machine. The manuals supplied aren't good enough by far, but the Digitalker chips are more versatile than you might think, so if you buy this one get in touch with National Semiconductor for the real data.

Of the remaining two units, the Cheetah offers a built-in amplifier and speaker whereas the Timedata unit doesn't; their respective prices reflect this. Neither of them comes close to the overall 'usableness' of the MicroSpeech and they both lack the BEEP amplifier and keyword voicing.

SWEET TALKER
Price: £29.75
Cheetah Marketing
24 Ray Street
London WC1R 3DJ

HARDWARE: Based on the by now familiar General Instruments allophone chip, this unit comes housed in a vertical box that measures 100mm by 75mm by 50mm.

Offering the benefits of an internal speaker (without volume control) and an expansion bus. Sweet Talker stays pretty much on a par with its rivals in terms of design; although it is well made and very neatly assembled on its single PCB.

SOFTWARE: This is the easy bit. Unlike the Currah MicroSpeech where the allophones are built up in strings (which is neat but consumes memory) the Sweet Talker simply uses numbers to generate the allophones. These can be stored as DATA statements ready assembled in word order, or you could build a list in alphabetical order sorted by position within a word. You can generate any of the allophones simply by keying OUT 7,n - where n is the allophone number.

MANUAL: It's virtually nonexistent and the meager four sides of A5 give you just the allophone list with some examples of how they would be used and that's about it. The tape includes a short demonstration which, according to the manual, "will explain precisely how to use it."

SUMMARY: The quality is as good as any other allophone-based synthesiser and the box does have the space-saving advantage of being vertically mounted.


REVIEW BY: Henry Budgett

Blurb: SPEAKER COMPARISON CHART Synthsiser: Currah Microspeech Synthesis Type: Allophone Allophone Coding: String Keyword Voicing: Yes Internal Amplifier: Uses TV Internal Speaker: Uses TV BEEP Amplifier: Yes Volume Control: Uses TV Demonstration Tape: Yes Software Provided: In ROM Games Available: Yes PHYSICAL NOTES Size (in mm): 75 by 75 by 28 Format: Horizontal Case Material: Plastic Expansion Bus: No Synthsiser: Fuller Box/Orator Synthesis Type: Allophone Allophone Coding: Numbers Keyword Voicing: No Internal Amplifier:Yes Internal Speaker: Yes BEEP Amplifier: Yes Volume Control: Yes Demonstration Tape: Yes Software Provided: No Games Available: Yes PHYSICAL NOTES Size (in mm): 235 by 100 by 48 Format: Horizontal Case Material: Plastic Expansion Bus: Yes Synthsiser: DCP S-Pack Synthesis Type: Compressed speech Allophone Coding: Numbers Keyword Voicing: No Internal Amplifier: Yes Internal Speaker: Yes BEEP Amplifier: No Volume Control: Yes Demonstration Tape: No Software Provided: No Games Available: No PHYSICAL NOTES Size (in mm): 75 by 110 by 45 Format: Horizontal Case Material: Plastic Expansion Bus: Yes Synthsiser: Timedata ZXS Synthesis Type: Allophone Allophone Coding: String/Numbers Keyword Voicing: No Internal Amplifier: No Internal Speaker: No BEEP Amplifier: No Volume Control: No Demonstration Tape: Yes Software Provided: On tape Games Available: No PHYSICAL NOTES Size (in mm): 65 by 78 by 40 Format: Upright Case Material: Plastic Expansion Bus: Yes Synthsiser: Cheetah Sweet Talker Synthesis Type: Allophone Allophone Coding: Numbers Keyword Voicing: No Internal Amplifier: Yes Internal Speaker: Yes BEEP Amplifier: No Volume Control: No Demonstration Tape: Yes Software Provided: No Games Available: No PHYSICAL NOTES Size (in mm): 110 by 75 by 50 Format: Upright Case Material: Plastic Expansion Bus: Yes

Transcript by Chris Bourne

Sinclair User Issue Annual 1984   page(s) 39,40,41,42

THE INGENUITY OF THE PERIPHERALS MANUFACTURERS HAS CONTINUED AND SINCLAIR RESEARCH FINALLY LAUNCHED THE MICRODRIVE. STEPHEN ADAMS REVIEWS ANOTHER INTERESTING YEAR FOR SINCLAIR USERS.

INGENUITY BEATS SPECTRUM LIMITS

Stephen Adams summarises the add-ons.

The biggest blow to users of non-Sinclair-produced equipment was the announcement of the Spectrum. It knocked most memory-mapped add-ons on the head, as no provision for those devices had been made. That was because, unlike the ZX-81, there was no way of turning-off the internal RAM from the expansion port. The only thing left was for producers to put things in the input/output map or to supply RAM packs or RAM upgrade kits.

The I/O map was already full of Sinclair devices, allowing the use of only eight separate devices, if they wanted them to be compatible with the Microdrive. Nevertheless, ingenuity brought forward a large number of new devices after the initial shock had worn off.

RAM kits for upgrading the 16K Spectrum to 32K are now readily available for about £21. Fox Electronics, for example, supplies a kit for £20.99 and it has instructions on how to improve your TV display. They are easy to fit and now have simple instructions on what to do, if you can overcome the fear of opening the case. The kits are usually identical to the proposed Sinclair upgrade but will fit only on model 2-type Spectrums. East London Robotics can also supply an extra 64K kit which is paged.

For those with a model 1 Spectrum, identified easily by the grey keys, there is the choice of an add-on board made by Downsway Electronics or the Spectrum 32K RAM pack by Cheetah Marketing. The Spectrum RAM pack will also fit on to any other Spectrum, as it plugs into the expansion port.

Another memory-mapped device available this year was in a surprising place, inside the Spectrum 16K ROM space. It was the Orme Electronics ROM containing RENUMBER, block delete of Basic lines and many other usefiil routines in a 2K ROM. That can be very useful, as it is immediately available on power-up.

Sinclair has also provided new devices in the ROM space, which should please hardware and software users. One is the Microdrive Interface One which not only contains the controlling ULA and hardware to run the RS232/network/highspeed cassette Microdrive but an extra ROM which can be used to write your own Basic commands. The other is the ROM cartridge system available with Interface Two.

The RS232 can handle a printer easily with its limited handshaking ability but Still is limited for input from an RS232 device, as all that is under software control and not a hardware chip. The network is a very simple arrangement which allows you to talk between 64 Spectrums but I have no doubt that someone will write similar software to use it with the ZX-81 through the cassette sockets. Two books which should help in this respect are those by Andrew Pennel and Dr Ian Logan.

The Pennel one also contains an ON ERROR GOTO machine code routine which should trap 90 percent of Basic errors and send them to a program line to deal with them.

Sinclair has also launched the Interface Two, containing two joystick sockets which will operate the first or last set of five number keys. That will mean that software will have to be re-written to use Sinclair joysticks, as the accepted standard at the moment is the one set earlier in the year by the Kempston joystick interface which operates as an I/O device, depending for its decoding on A6 only being low.

Much software has already been converted to use it and I cannot see software writers wanting to do it all again. Programmable joysticks are just starting to reach the market, which will eliminate the need to rewrite the software.

For the Spectrum and soon the ZX-81, the Stonechip one seems to be best. For ZX-81 and Spectrum users, a cheaper hardware-based version, the Pickard controller, can be used as it fits both computers by plugging into the keyboard sockets. The AGF version uses crocodile clips to set up the five keys to be used; it is a little unstable but it plugs into the edge connector and does not require entering the machine.

The Interface Two also has a single ROM socket which allows you to use Sinclair-designed cartridges. It is not known whether software suppliers will want to try to fit normal ROMs to a port like this. That, and the fact that recording Microdrive cartridges is a slow process, will limit the amount of software available from software manufacturers on Sinclair devices. EPROM software for ZX-81s is available from Eprom Services and Audio Computers.

Both, however, concentrate on the machine code user and not the games player. The fall in price of the ZX-81 and the amount of hardware available for it has started to attract the business user as a control processor for a robot or controlling some industrial process.

That is because it contains all the requirements of the development system on one board - Z-80-A, one of the most popular processors, working at 3.25MHz, TV interface, cassette interface for program storage, alphanumeric keyboard, Basic as well as machine code monitor for designing programs, and an expandable memory map and I/O map.

There is also a wide range of machine code assemblers, disassemblers, EPROM cards and blowers, RAM, battery-backed memory already available, costing just a few pounds. That to the industrial process manufacturer is peanuts, as a development kit from a chip manufacturer would cost more than £200 with far fewer facilities. The ZX-81 costs only £45 with a massive 16K memory included.

Plain-paper printer interfaces for the Spectrum abound and Tasword, the word processing program, can work with most of them. Hilderbay and Kempston provide software-driven Centronics versions, while Morex provides both RS232 and Centronics outputs. Deans also introduced its version of the Timex printer, which uses a much better paper than the Sinclair, gives a clearer print in both black and blue ink, and will use Sinclair commands to control it, so there is no need for extra software.

Printers and other hardware devices make the software "come alive" and that combination makes the Spectrum a very powerful business computer.

That is the most important event of the year, as the more software which is written to use the large amount of hardware, the more that hardware will be used. Voice output units, modems - to talk to other users or databases over the telephone - RS232 interfaces and light pens all rely on good software to make use of them.

The Cheetah Marketing Sweet Talker, for instance, works better because it is accompanied by an instruction tape which not only demonstrates how to use the unit but also shows the user how to structure programs to make it easier to use in their programs.

Modems from Maplin, Ambit and Micronet - for Prestel - will allow users of the Spectrum and, in the first two cases, the ZX-81, to talk to many other computers. Some of them will be other types of computers, like the BBC and the Commodore 64. The others will be maintained by public and private companies which maintain large amounts of information and programs on their computers.

Micronet 800 is a database maintained on Prestel computers as a 24-hour-a-day, seven-day-a-week computer club. It has hints and tips on Spectrums - ZX-81s at the moment cannot use the system - as well as news, free programs and a mail box facility. Keyboards and consoles also have been making their presence felt as ZX-81 and Spectrum users want to upgrade their machines. Plastic cases from W H Smith are about the cheapest containers at £3.99 and will take a ZX-81 or Spectrum, as well as a few add-ons. The d'Ktronics keyboard and case or that from Fullers appear to be the most popular but neither will allow the use of the Microdrive interface without taking it out of its case.

The Filesixty button set is a cheap alternative for ZX-81 users which gives the advantage of spring-loaded keyboard the same size as that of Sinclair, but without costing more than £10.

Colour for the ZX~81 is now available in a simple form for all PAL television users - that is, most of Europe. The black box requires only two wires to insert between the modulator and the breaking of tracks and can be re-connected if required. The rest of the box plugs into the back of the ZX-81 to give black characters on a choice of 16 coloured backgrounds or coloured characters on a black background from DDC. It is also more stable than the Spectrum and requires no extra memory to use it. It will work even on a 1K machine.

One device which requires extra memory is the excellent High-res screen - 192 by 256 pixels - and user-definable graphics package made by Nottingdale Technology Centre. It uses the ZX-81 internal 1K RAM for its system variables and 6K of memory for the storage of the screen in the program. No internal wiring is required, as the unit plugs directly on to the back of the ZX-81.

For the Spectrum, a unit has appeared which will be of great delight to children and disabled users. It is the Currah microSpeech unit which plugs into the back of the Spectrum and, on command, will speak the key pressed. That happens during program input as well as INPUT and INKEY$.

It requires no programming of the speech by the user. Speech output can also be programmed by using S$ to contain the allophones - sounds which make up words - which are then spoken immediately. Thus input and output can be spoken rather than read.

Tapes have been causing problems ever since the ZX computers came into existence; the Spectrum is better than the ZX-81 but still can be improved with some extra hardware.

Tape filters and switch-controlled SAVE and LOAD devices are available from several firms, like Abacus and Elinca. There have now been three Spectrums produced by Sinclair. The model is, which can be identified by the grey keys or by looking through the expansion interface and seeing an IC socket on the left-hand side; the model 25 have the large, black, ROM chip there. Those Spectrums need their extra 32K of RAM mounted on a printed circuit board before they can be put into the computer. There are no Sinclair RAM boards available for the machine.

The model 25 were re-designed completely by a computer and the ULA was changed to get rid of an extra 1C which had to be inserted in the model is, due to a design error. The extra 32K of RAM which can be added to 16K machine now requires only chips to be plugged into sockets on the board. Model 1s and model 2s have had to have an extra transistor fitted to prevent a clash between the keyboard and the ULA TV interface.

The latest model 3s have also had their internal circuitry re-arranged and the ULA updated to give a wider tuning range on TV sets. It has also caused some software problems, as the keyboard inputs are no longer held to binary l - + 5 volts - when not in use. That was done to reduce the power 1 requirements of the ULA.


REVIEW BY: Stephen Adams

Blurb: 'Printers and other hardware devices make the software come alive.'

Transcript by Chris Bourne

All information in this page is provided by ZXSR instead of ZXDB