ASR Engine Overview

The Vestec Speech Engine has been designed as a powerful, low-cost, standards-based alternative to conventional ASRs for small vocabulary continuous speech applications. These applications constitute the majority of “command-and-control” type IVR deployments in the world.

Our ASR engine offers powerful capabilities at affordable prices, including:

• SRGS (both ABNF & XML) grammar format
• MRCP (both v1 & v2) platform integration
• Acoustic models for 4 major languages
• High accuracy (for native & non-native speakers)
• Scalable architecture and load balancing
• Advanced echo cancellation & noise reduction
• Soft-PBX Asterisk and FreeSWITCH support

The ASR engine is offered in two models. The Tier-1 engine supports a vocabulary size of 500 items per recognition and retails for $199 per channel. The Tier-2 engine supports a vocabulary size of 2,500 items per recognition and retails for $299 per channel. Acoustic models need to be licensed separately at $20 per language per channel.

Retail prices include installation support as well as first-year software maintenance. From 2nd year onwards, annual software maintenance (covering bug fixes and upgrades) can be purchased for Tier-1 and Tier-2 engines for $29 and $49 per channel, respectively. Non-installation support – including, product training, grammar writing, and system integration – is also available at competitive rates.