Parsippany, New Jersey-based First Byte Inc, specialist in text-to-speech synthesis technology, and Dialogic Corp, developer of telephony-based speech boards for personal computers, have a new version of ProVoice that supports Dialogic voice boards, a software tool kit for programmers to add human-sounding, synthesised speech to MS-DOS-based applications. The second phase converts the intermediate phonetic language into speech signals, and algorithms drive distinct speech signals into smoothly flowing, continuous clear speech, the firms claim. ProVoice for Dialogic costs $125 per port plus a one-time development kit licensing fee of $1,200. Licences are available for two, four and eight ports. It will be available for OS/2 by the end of the year. And Dialogic has two universal call technology products based on its new D/41E four channel voice board. They provide four channels of integrated voice and facsimile or voice and speech recognition in a single AT expansion slot. The VFX/40 incorporates facsimile processing for development of integrated speech and facsimile response systems. The VSR/40E integrates automatic speech recognition for development of voice response systems that can accept spoken commands in place of touch-tone input.
Unified messaging
They can be used to create systems for applications like unified messaging, voice and facsimile response, and systems that can be accessed from rotary dial telephones. Supporting the new Signal Computing System Architecture and Dialogic’s PCM Expansion Bus and Analogue Expansion Bus architectures, the VFX/40 and VSR/40 are more scalable, enabling developers to use the same products to build low-density to high-density systems. Based on technology from Voice Control Systems, the VSR/40E uses advanced algorithms to recognise spoken commands for call processing applications. Standard features include 15-word speaker-independent recognition in over 25 languages and Voice Cut-Thru technology that enables users to speak commands at any time during message playback. With Voice Cut-Thru, rotary phone users can bypass a pre-recorded message in the same way touch tone users type ahead to interrupt menu prompts, for new applications like hands-free voice messaging. The VSR/40E also supports the full set of VCS Enhanced Speech Recognition Technologies, including alphanumeric and speaker-dependent recognition, speaker-verification and custom vocabulary creation. The D/41E, VFX/40 and VSR/40E support multiple call processing architectures, enabling them to be used in low density systems with scalability to high-density configurations. These products support Dialogic’s Analogue Expansion Bus for low density systems and can be upgraded for Signal Computing Systems Architecture or PCM Expansion Bus compatibility for systems requiring higher-bandwidth resource sharing. Production quantities of VFX/40 will be available this month for $2,110 in volume. The VSR/40E costs $4,160 in volume, and is out in early 1994. Both products will be supported with development packages for the MS-DOS, Unix and OS/2 operating systems.
