Voice interface is always-on, on-device, at the edge
Edge AI processor company Syntiant Corp. (Irvine, Calif.) has announced a proprietary on-device, small language model voice assistant (SLMA).
The voice recognizing software is portable across hardware platforms and offers a natural voice interface for human-machine interactions. It is suitable for set-top box manufacturers, service providers and other consumer electronics device manufacturers. It allows much of the power of the large language model (LLM) to be brought to the edge without needing to significantly upgrade or replace existing hardware, Syntiant claims.
“With our technology, AI language models that typically demand extensive computing resources can now operate efficiently on local devices,” said Kurt Busch, CEO of Syntiant, in a statement. “While voice applications for our SLMA extend across many industries, imagine the advantage for set-top box manufacturers. Instead of providing static guides or instructions, manufacturers can now integrate a more dynamic, conversational AI assistant to help users without a Web connection, reducing one of the main customer friction points for service providers.”
The SLMA is being by early adopters for voice interfaces in home appliances, networking equipment and video conferencing systems. Coupled with Syntiant’s automatic speech recognition models, Syntiant’s SLMA provides a natural voice-interface at a fraction of the computational cost of LLMs.
SLMA has 23 million parameters – compared to the billions used within broad application LLMs – and as a result can be run on most set-top box CPUs. The SLMA has always-on capability.
Synaptics has integrated the SLMA on to its Astra SL1680 SoC.
Related links and articles:
News articles:
CEO interview: with Kurt Busch of ‘always-on’ startup Syntiant
Agile Analog delivers configurable ‘always-on’ IP to XMOS
Event-based sensor for ‘always-on’ video, low-power applications
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
