Loquendo is now part of Nuance.  Learn more about Nuance's speech recognition products and solutions >

Loquendo ASR Voice Recognition Software – Distinguishing Features

Loquendo ASR Voice Recognition technology offers an unparalleled range of advanced features that makes it stand out in terms of performance, versatility and functionality.

Loquendo ASR:

• combines Neural Networks and Markovian Models, to reduce memory requirements and achieve high levels of accuracy even on large vocabularies 
• is speaker independent, but is equipped with a tool for adapting acoustic models with data collected in the field, if required by a particular application
• offers several product profiles optimized for the various fields of use: Telephony, Mobile, Automotive, Embedded
• is particularly robust to background noise and speech, thanks to a denoising algorithm
• supports voice recognition based on both grammars and statistical language models
• offers Keyword Spotting, for rapid content analysis.
• can be equipped with a semantic classifier for creating automatic call routing applications based on caller requests 
• provides re-usable built-in libraries for recognizing, for example, addresses, currency values and alphanumeric strings
• supports speaker verification
• supplies information on phonetic segmentation, for the animation of digital assistants (avatars).

For a complete description of the product’s technical features, consult the Speech Recogntion Product Brochure.

Standards-Support to Safeguard Your Investments

Loquendo Voice Recognition software conforms to speech standards developed by the W3C (World Wide Web Consortium), of which Loquendo is an active member:

• Speech Recognition Grammar Specification - SRGS 1.0 (XML & ABNF formats)
• Semantic Interpretation for Speech Recognition - SISR 1.0
• Pronunciation Lexicon Specification - PLS 1.0 
• Extensible MultiModal Annotation markup language - EMMA 1.0
• Media Resource Control Protocol - MRCP v1 & v2 (via Loquendo MRCP Server).