Stunningly accurate, embedded and cloud-based voice recognition.
Cerence ASR (Automatic Speech Recognition) is a unique SDK for embedded and connected speech recognition applications. This feature-rich development suite supports creating voice-based interfaces for cars, home appliances, logistics, smart homes, and many more. Natural language understanding and speech signal enhancement are available to create robust and intuitive voice control solutions.
This advanced speech recognition engine delivers a new level of speaker-independent and continuous speech recognition capabilities with unique features. It provides superior functionality, unmatched accuracy, and high performance for a variety of applications that benefit from speech control.
Designed as a modular and scalable engine, Cerence ASR can accommodate a large range of embedded platforms. Deployments can be customized with optimized footprints as dictated by the required functionality for each application.
Features and Benefits
Cerence ASR is optimized for a wide variety of applications and has been deployed in automotive, accessibility, industrial, and consumer electronics solutions for over 15 years.
Modular and Scalable Architecture
Supports a range of operating points so that smaller applications are not affected by higher CPU & RAM required for larger applications.
Tools for Developers
Includes sample applications and all tools required to build speech recognition applications.
Support for a Large Vocabulary in Many Languages
Enables embedded recognition for large lists up to millions of items in more than 30 languages.
Exceptional Performance across Range of Environments
Cerence ASR is developed for mobile applications with high accuracy operation to SNR levels as low as SdB.
Bolstered by the Broad Cerence Portfolio
Cerence ASR is part of an unrivaled portfolio of speech and language solutions, working on its own or as a fine-tuned chain of technologies, including noise reduction, speech recognition, language understanding, dialogue, text to speech, voice biometrics, and more.
Enables all commands from all domains to be spoken in a single utterance on the main menu.
Wake Up Word Ready
Always listening mode with key-word activation removes the need for a “press to talk” button.
Freedom of Message Dictation
Recognizes free-form dictation text, together with a message command, all in a single utterance.
Flexible Points-of-Interest Recognition
Recognizes every possible word in a database in every possible order, especially suited for Points of Interests (POI).
Broad Domain Support
Dedicated solutions for key domains, such as music, destination entry, telephony, and more. Multilingual recognition (e.g. for domains, such as music and destination entry).