Products + Solutions
DynaSpeak® Speech Recognition Engine
DynaSpeak is a small-footprint, high-accuracy, speaker-independent speech recognition engine. It scales from embedded to large-scale system use in industrial, consumer, and military products and systems. The technology is available for license from SRI.
The speech engine incorporates SRI-developed, patented techniques that increase recognition performance using speaker adaptation, microphone adaptation, end of speech detection, distributed speech recognition, and noise robustness.
DynaSpeak is a product of SRI’s Speech Technology and Research Laboratory, a world leader in speech technology research and development.
Key Features
- Small footprint (750KB–2MB); enables compact deployment
- Statistical grammar option for natural language and domain-specific dictation
- Adaption to background noise
- Compact grammars for fast updates
- Speaker adaption improves recognition with each new utterance
- Non-native speech models for robustness to strong accents
- Supports very large vocabulary (50,000+ words)
Specifications
Minimum processor requirements
- 200 MHz StrongArm
- 66 MHz Intel x86
- Support for other hardware available on request
Minimum memory requirements (KBs)
- Total: 750
- Executable (ROM): 350
- Acoustic models (ROM): 100
- Active mearch (RAM): 300 or greater for highly complex grammars
Supported languages
- Adults: American and British English, Latin American Spanish, Iraqi Arabic, Pashto, and Dari (others on request)
- Children: American English
- Grammars: Statistical or JSGF forms; static or dynamic; dictation option
Software environments
- Operating systems: Windows, Mac OS X, Linux, and Android (others on request)
- Development: C/C++, Java (via JNI), client/server versions available
