ACUVOlCE's SPEECH SYNTHESIZER BRINGS TEXT TO IIFE
AcuVoice's (San Jose, CA--408-289-1661) AV1700 text-to-speech synthesizer translates any ASCII text into speech -amazingly human-sounding speech.
Linguist E. David Barton, president of AcuVoice, was on hand to enthusiastically explain the inner workings of the AV1700. Barton claims the product focuses around two questions: What is the phenomenon of human speech and what is the phsiology of emitting sounds?
One you understand the science surrounding those two questions, it's a piece of cake to create a neural net-based software product that emulates the mechanics of the human mind in the complex process of reading and voicing text. So says Barton; and who the hell am I to argue?
Barton's been developing the technology since the early 1960s; he started AcuVoice in 1986. Because the product doesn't store words, but stores almost 100,000 human utterances, Barton's first chore was to define each phonetic syllable in the English language--a language with over 700,000 words. Cunning.
It works by using a concatenative process to analyze the syntax created by a sel ies of words, then determine how each word must be pronounced to bring out the intended meaning of the text being processed. The system then replicates how humans convert a line of text into a stream of text-sensitive sound pulsations.
Meantime, on the all important what-can-it-do-for you front:
The AV1700 speaks e-mail, memos, text files and lists of numbers, dollar amounts, dates and greetings with spooky naturalness. Because it frees the eyes and hands to perform other tasks, the program is useful in several industries, from education, office automation, manufac- turing, transportation, medicine and for speech-impeded individuals. Cost for the AV1700 Text Reader is $99.
To run the software you need an IBM compatible PC with Pentium processor carrying a minimum of 75 MHz, Windows 95 or NT, 8 Mb of RAM, a minimum of 160 MB on your hard drive, a CD-ROM drive and any sound card supporting Mu Law PCM Wave Files (such as Sound Blaster 16 or Port-Able Sound Plus) or Linear PCM Wave Files (such as Sound Blaster Pro).
AcuVoice also has the AV2001 Telephony/Multimedia Interface Developer's Kit for UNIX, Solaris and WIN 32, so you can integrate the system into your own telephony and multimedia systems. Barton claims that AcuVoice's market is 75% telecommunications, with 15% in multimedia.
That 15% is the market to watch, he says, as CPU Power and disk storage not only on the desktop but also on hand-held devices become more of an issue. Acu- Voice's Speech Synthesizer AV2001 telephony developer kits from one channel to Solaris Sparc are avail- able from $900 to $3,600.
AcuVoice had other news to broadcast from its booth as well. The 12-year-old company has re- cently been acquired by Fonix (Salt Lake City, UT 801-328-0161), a com- pany most concerned with using speech to interface between comput- ers and humans.
file: /Techref/app/accuvoice.htm, 3KB, , updated: 2002/1/14 15:27, local time: 2024/12/2 08:02,
3.145.57.41:LOG IN
|
©2024 These pages are served without commercial sponsorship. (No popup ads, etc...).Bandwidth abuse increases hosting cost forcing sponsorship or shutdown. This server aggressively defends against automated copying for any reason including offline viewing, duplication, etc... Please respect this requirement and DO NOT RIP THIS SITE. Questions? <A HREF="http://massmind.org/Techref/app/accuvoice.htm"> app accuvoice</A> |
Did you find what you needed? |
Welcome to massmind.org! |
Welcome to massmind.org! |
.