LumenVox announces v5.5 of three speech tools.

LumenVox has announced version 5.5 of its Speech Recognition Engine (SRE), LV Speech Tuner and Speech Driven Information System (SDIS). The latest version of the SRE offers improved performance, with the introduction of lattice-based confidence scores. LumenVox has also introduced improved barge-in and/or End-of-Speech (EOS) detection, which can detect when the caller has begun speaking, finished speaking or paused while speaking. LumenVox allows developers to custom-design their own grammars on-the-fly, filter out-of-vocabulary words and modify the engine and grammar performance at runtime. Built-in grammars handle single digits, currency, natural numbers, and date and time.

Version 5.5 of the LV Speech Tuner (LVST) is a maintenance tool that allows end users, value-added resellers and platform providers to perform tuning, transcription and testing of any speech-driven application created on multiple ASR platforms. Version 5.5 of the LVST enables speech application tuning, parameter testing, grammar testing, version upgrade testing and transcription. With this GUI-based tool, companies developing speech applications on various ASR platforms can bring speech application tuning in-house and avoid professional service fees.

The latest addition to the Speech Driven Information System (SDIS) is support for a VB ActiveX exe interface. This interface can now allow developers to program their required application logic with the programming language Visual Basic. Developers can utilize common prebuilt libraries to access databases and premade integration components. This gives speech application developers the option to extend their SDIS projects with either C/C++ or Visual Basic.

www.lumenvox.com