SoftVoice, Inc. is pleased to announce the availability of its new, fifth generation, multi-lingual text-to-speech system. The SoftVoice text-to-speech system (SVTTS) converts unrestricted English and/or Spanish text into highly intelligible, natural sounding speech in real time. SVTTS uses no recorded speech, so it has an unlimited vocabulary. Explore this page and see why the world's major computer and software companies have selected SVTTS for their applications.

About SoftVoice, Inc.
About the SoftVoice text-to-speech system (SVTTS)
Licensing SVTTS for inclusion in your software product
Download SVTTS audio examples


Clicking on any will pop you back to this menu.

About SoftVoice, Inc.

SoftVoice, Inc. has over 25 years of experience in speech synthesis. The first, commercially available, all-software text-to-speech synthesizer for microcomputers was written by the people at SoftVoice in 1979. S.A.M. (Software Automatic Mouth) was a best-seller on Apple, Atari, and Commodore computers.

In the early '80s, SoftVoice developed text-to-speech systems for both Apple and Amiga Corporations for their soon-to-be-released machines. In fact, at its launch in 1984, the Macintosh© announced its own existence to the public using our software, the original "MacinTalk"©. The Commodore Amiga had an early SoftVoice text-to-speech system known as the "narrator.device" included in its operating system.

To date, there are over 8 million copies of the SoftVoice text-to-speech system in use world-wide, making SoftVoice, Inc. one of the largest providers of text-to-speech in the world.

But that's all old news. Read about our new, fifth generation product, SVTTS for Windows©.

The SoftVoice Text-to-Speech System

Implemented as Windows© DLL's, SoftVoice TTS is a state-of-the-art expert system for the conversion of unrestricted English text to high quality speech in real time. The SoftVoice system is the most versatile in the industry, with 20 different preset voices that the user can quickly select according to his or her preference. The programmer or user can also alter any of the voices in virtually limitless ways to create totally new voices. Of course, the usual parameters such as pitch and speaking rate are variable over a very wide range. The user can also choose between normal, breathy, or whispered vocal qualities. A comprehensive set of over 30 different commands can be embedded in the text to control the speech output. SoftVoice even has extensive singing support!

During speech, the programmer can elect to receive a variety of different messages back from the synthesizer. Program events can be synchronized to speech events such as start-of-word, start-of-syllable, or user defined locations. Also, accurate mouth shapes can be animated from data supplied by the synthesizer.

The SoftVoice system utilizes letter-to-sound rules, a numeric pre-processor, and a dictionary to determine proper word pronunciation. Programmers or users can also use the SoftVoice exceptions dictionary editor to create their own dictionaries for words or abbreviations not pronounced properly.

SVTTS is multi-lingual with English and/or Spanish currently available. Through the use of embedded commands, the language can be switched within a sentence. For example, you can write:

"{{spanish}} Tiene un perro loco, {{english}} means you have a crazy dog."

More languages are under development and will be released in the near future.

The SoftVoice system is built around the concept of formant synthesis in which we mathematically model the human speech production mechanism and, in particular, the acoustic resonances (formants) of the vocal tract. As opposed to the time-domain methods (demi-syllable, diphone, etc.) in which pre-recorded pieces of speech are spliced together, formant synthesis is not restricted to a single sounding (or minimally modifiable) vocal personality, nor is it constrained to a narrow pitch range. With the SoftVoice system, users can customize vocal personalities to suit their individual needs, everything from a singing choir boy to a menacing Martian. In addition, the SoftVoice formant synthesis algorithm, being continuous and splice free, does not suffer from such artifacts as glitches, gurgling, false consonants, chorusing, etc. that reduce intelligibility and increase listener fatigue.


The following is a list of some of the more prominent features of the SoftVoice text-to-speech system:

• Real time conversion of unrestricted text to high quality speech
• Available in English and/or Spanish. More languages under development
• Twenty pre-defined vocal personalities including males, females, children, robots, and aliens
• An unlimited number of vocal qualities/personalities can be created by the user
• Large selection of embedded commands control myriad parameters
• Speaking rate control, very wide range (20 - 800 words per minute)
• Pitch control, very wide range (10 - 2000 Hz)
• Pitch modulation effects (vibrato, perturbation, excursion, etc.)
• Control of vocal quality (normal | breathy | whispered)
• Control of speaking mode (natural | word-at-a-time | spelled)
• Can return mouth shape data for use in animation
• Synchronization capabilities (sync messages can be sent to the user's application signalling a variety of events)

Licensing SoftVoice TTS

If you are developing a commercial or industrial software product, you can license the SoftVoice text-to-speech system for inclusion. Licensing of the SoftVoice TTS engine can be done in a number of ways, including (but not limited to):
   - A per-unit royalty with large-volume discounts, or
   - A yearly subscription, or
   - A single, one-time fee.

For information on licensing the SoftVoice TTS engine - or for general questions - please contact us at: SoftVoice Information Request for details.

Download Audio Examples

Here is a small sample of the fun you can have with the SoftVoice text-to-speech system. With 20 modifiable voices, and over 30 types of embedded commands that can be inserted into your text, the possibilities are limitless. The examples below are .wav files recorded at 11KHz, 8-bit mono, and should play on any Windows© machine equipped with a sound card.

Download audio example "We have all the flavors, as long as you want chocolate, vanilla, or strawberry."
Spoken by Standard Male.     Size = 59KB

Download audio example "This is the voice of world control. Obey me and live, or disobey and die."
Spoken by Colossus.    Size = 66KB

Download audio example "Twinkle, Twinkle Little Star."
Sung by Child.    Size = 67KB

Download audio example Identity Crisis!
Spoken by various SoftVoice personalities.    Size = 60KB