Speech, Language, and Audio

From soothing a crying baby to rallying millions to a cause, the human voice is a powerful tool. Speech, language, and audio also allow us to interact naturally with a variety of computing applications.

For government and commercial clients, SRI's Speech Technology and Research (STAR) Laboratory creates speech recognition software used for signal processing, phonetics, mathematical modeling, acoustics, data indexing and mining, two-way language translation, and computer-aided learning. For tailored client solutions, SRI offers advanced product prototyping and customization and turnkey product development.

SRI also develops speech recognition technology for use in challenging audio environments, where hearing a particular speaker is difficult. SRI has honed speaker identification and verification techniques for biometric applications.

Developers incorporate SRI’s DynaSpeak® and EduSpeak® technologies, packaged as software development kits, into their products and services.

Market leader Nuance, now a billion-plus-dollar company, was spun off from SRI in the 1990s to exploit STAR Laboratory technology. Today, developers incorporate SRI speech technologies, packaged as software development kits, into their products and services.

Projects

U.S. soldier shaking hands with an Afghan man
Broad Operational Language Technology (BOLT) Program

SRI is creating technology to translate multiple foreign languages in all genres, retrieve information from translated material, and enable bilingual communication via speech or text.

Topics: Force Protection Intelligence, Surveillance + Reconnaissance Speech, Language, and Audio

Products + Solutions

DynaSpeak logo
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.

Topics: Speech, Language, and Audio
EduSpeak logo
EduSpeak® Speech Recognition Toolkit

EduSpeak is a speech recognition toolkit specifically designed for developers of language-learning applications (such as for English as a Second Language, or ESL) and other educational and training software. The technology is available for license from SRI.

Topics: Speech, Language, and Audio
people in a call center
SRI Language Modeling Toolkit

The SRI Language Modeling (SRILM) toolkit offers tools for building and applying statistical language models (LMs) for use in speech recognition, statistical tagging and segmentation, and machine translation. The toolkit can be downloaded and used free of charge (more information below).

Topics: Speech, Language, and Audio

Press Releases

Soldier and civilian
DARPA Awards BOLT Program $7.1 Million Contract to SRI-led Team

SRI International has been awarded a $7.1 million contract for Phase 1 of a five-year, $41.5 million Defense Advanced Research Projects Agency (DARPA) contract under DARPA's Broad Operational Language Translation (BOLT) program.

Topics: Force Protection Intelligence, Surveillance + Reconnaissance Speech, Language, and Audio
woman talks on mobile phone
DARPA Awards Contract to Develop Robust Speech Processing Software for Noisy Environments

SRI International has been awarded a $13M contract from the Defense Advanced Research Projects Agency (DARPA) to develop software to process noisy and highly degraded speech.

Topics: Speech, Language, and Audio
SRI International and SYSTRAN Announce Licensing Agreement for SRI’s Language Modeling Toolkit

SRI International, an independent nonprofit research and development institute, and SYSTRAN, a leading provider of language translation technologies, today announced that SYSTRAN has licensed and fully integrated SRI’s Language Modeling (SRILM) toolkit.

Topics: Speech, Language, and Audio
SRI International Licenses DynaSpeak Speech Recognition Software to Adacel for Aviation Applications

Adacel, an industry leader in simulation development, software integration, and speech recognition technology, has obtained an exclusive license for SRI International’s DynaSpeak® speech recognition system for selected aviation voice-activated cockpit and mission specialist applications.

Topics: Speech, Language, and Audio
SRI International Offers Patent Portfolio for Voice-to-Database Search Technology

SRI International is offering for sale its voice-to-database (V2DB) search patent portfolio, which addresses methods for spoken search and retrieval of information from large databases using a speech recognizer. The methodology can be applied in various markets.

Topics: Speech, Language, and Audio
SRI International Receives Patent for Voice-Based Search Technology

SRI International has been issued a second U.S. patent for its voice-to-database (V2DB) technology that enables one-step, voice-based access to items in vast, multimillion-record databases. SRI's one-step approach can work with currently available speech recognition systems.

Topics: Speech, Language, and Audio
SRI Researcher Receives International Speech Communication Association (ISCA) Fellow Award

SRI Researcher Receives International Speech Communication Association (ISCA) Fellow Award

Topics: Speech, Language, and Audio

SRI In the News

DARPA Program to Power Instant Translation of Multi-Lingual Email, Messaging and Speech

This article reports that “The Defense Advanced Research Projects Agency awarded a $7.1 million contract to SRI International to start building the latest in a long line of technologies that seek to translate and understand multiple languages.”

Topics: Artificial Intelligence Speech, Language, and Audio
F-35 Pilot-Aircraft Speech System Fine-Tuned

This article reports that SRI International developed DynaSpeak speech recognition software as a highly accurate system for noisy environments, specifically for embedded devices like personal digital assistants, in-car navigation systems and avionics systems.

Topics: Speech, Language, and Audio
Desti logo
Siri’s Sister Desti is a Virtual Tour Guide to Answer All Your Travel Queries

This article reports that Desti is not just another travel app. It is a virtual tour guide that uses artificial intelligence and natural language processing to comprehend and respond to specific queries. The startup launched its open beta today with technology developed out of SRI.

Topics: Artificial Intelligence Speech, Language, and Audio
SRI International Gets $13M from DARPA for Distorted Speech Software

This article reports that SRI will work on software to process "noisy and highly degraded speech" under a $13 million contract from the from the Defense Advanced Research Projects Agency (DARPA).

Topics: Speech, Language, and Audio

Publications

Speaker Recognition with Region-constrained MLLR Transforms

In this paper we ask if another major class of speaker recognition models, those based on MLLR speaker adaptation transforms, can also benefit from region-constrained feature extraction.

Topics: Speech, Language, and Audio
Meet Our Innovators: Colleen Richey

More Videos (3)

Divisions, Centers + Labs