Main menu

Automatic Speech Recognition (ASR)

Automatic Speech Recognition

In making interactive applications more intelligent, accurate automatic speech recognition capabilities enhance user-experience.

GoVivace offers a reliable and robust Automatic Speech Recognition engine. We provide the speech recognition software as a Software Developer Kit (SDK) library as well as the websocket with bidirectional streaming to use the software as a service. These include interfaces for Independent Software Vendors and developers, even those working on cloud-based applications.The speech to text engine is designed to work 24 X 7 X 365 and can be used to build mobile, web as well as high volume application.

GoVivace's Automatic Speech Recognition engine can accurately recognize spoken words and convert speech into text. It supports several English accents and can be localized to any language. Also, it supports standard telephony as well as web and mobile applications. Being capable of actioning voice commands given to electronic devices such as computers, tablets, smartphones or telephones, the GoVivace Automatic Speech Recognition engine finds use in diverse applications.

The engine compares the spoken input with a number of pre-specified possibilities. The entire set of pre-specified possibilities constitute the application's grammar, which powers the interface between the dialogue-speaker and the back-end processing. The GoVivace speech recognition solution needs only very simple grammar, but also supports very large grammars for complex tasks such as dates, complex commands and yellow pages styled complex directory lookups. GoVivace additionally offers consulting services for the construction of complex grammars. Performance tuning is another service, whereby GoVivace troubleshoots poorly performing grammars.

The GoVivace's speech recognition software can work with both pre-compiled grammars that can be referenced by name, and on-the-fly grammars that evolve as the client uses the application and which can be detected if reused. Both kinds of grammars are stored on the server after compilation, to ensure fast processing.

A key feature of the GoVivace Automatic Speech Recognition engine is it uses a statistical language model to understand natural language, which means it is not limited to understanding speech which matches its grammar. Developers integrating the engine into speech recognition systems will appreciate being able to create advanced intuitive natural language processing interfaces boasting of high linguistic intelligence quotients.

GoVivace speech recognition software is available in 32 and 64 bit Linux, Windows and Mac versions, for enterprise and SMB customers.

The GoVivace automatic speech recognition solution supports a distributed client / server architecture for easy scaling and to support an ever growing list of client devices. A load balancer can be used as the front end, and servers added to the system at the back end to allow for redundancy, reliability and scalability.

The GoVivace Automatic Speech Recognition is available in both 32 and 64 bit versions for Linux, Windows and Mac platforms. A minimum of 4GB of RAM and 2.0GHz processor is recommended.