Name Pronunciation Software. We have released version 3 of the E-Speech name pronunciation software, which includes significant improvements to US geographical names, as well as to people's names. Our latest software also offers alternate pronunciations for many names that have multiple pronunciations. For example, "Xavier" can pronounced as either "ehks A vyer" or as Spanish "ha VYER". A speech recognition system must be able to link both pronunciations to the same spelling. We provide software or dictionaries that can be used in your speech recognition or text-to-speech systems, as well as in systems with live call-center agents. E-Speech offers the most accurate pronunciation software available today for the names of people, places, and businesses in the United States.

Spanish Synthesis. E-Speech has introduced a Latin-American Spanish synthesis system that speaks people's names with extremely high sound quality. The system converts orthography to phonemes and stress marks and, based on E-Speech's proprietary subword concatenation algorithms, generates speech from its sound inventory. This system illustrates the sound quality that our custom speech generation systems can offer you for domain-specific applications such as names, addresses, timetables, financial transactions.

NSF grant awarded to E-Speech. E-Speech is proud to announce that we have been awarded a 2004 National Science Foundation Small Business Innovation Research (SBIR) grant to develop software to teach reading. Our software is based on E-Speech's proprietary letter-to-phoneme rules.

(see more news)