TalkTo Audio Front End for Voice Controlled Products

TalkTo™ is a suite of proprietary Machine Learning & Microphone Processing algorithms that delivers a quiet-room experience in the noisy & unpredictable spaces of everyday life.

 

 

 

 

To support the full spectrum of voice-enabled products, TalkTo is available in multiple variants, each scaled and optimized for the needs and capabilities of the target application.

 

 

Smart Speaker

 

Set Top Box

Battery Optimized

 

Hearable / Wearable

 

TV / Smart Screen

Multi-channel 

Smart Home / IoT

 

 

 

 

 

Anatomy of a Voice-Controlled Product

 

 
 

Audio Front End (AFE)

Conceptually, the AFE serves as a microphone cleaner. It uses the raw, noisy audio from the microphones to detect, extract, and clean any speech-activity found in the ambient sound field. 

Wake Word Engine (WWE)

Wake-Word Engines continuously scan the AFE-provided voice-stream, looking for the presence of a specific phrase (e.g. "Alexa").  Accuracy of the WWE is a primary driver of UX-quality and depends heavily on AFE quality.

Back End Services

After the wake-word is detected, the extracted voice-command is passed to a [cloud-based or local] voice-service which converts speech-to-text (ASR) and determines user-intent (NLU). Again, accuracy is highly dependent on AFE performance.

Speaker Processing

For robust Voice UI performance, products with audio-output(s) must have low speaker-distortion. To achieve this, output-audio is digitally limited to the physical capabilities of the speaker(s). Audio Weaver® is a fast path solution to this problem.

 

 

 

 

 

TalkTo Performance

Thanks to DSP Concepts' exclusive Adaptive Interference Canceller™ (AIC), TalkTo™ is able to detect and extract faint voice-commands, even in noise-saturated environments.

 

The video below shows the 6-Microphone, Amazon Qualified, TalkTo - Set Top Box  variant running on an Amlogic A113 board. Note that AEC can't be used in this and many other use cases, as the voice-enabled device has no knowledge about the noise source (here, the TV speakers).
 
 

 

 

For more information or to schedule a live demo, please contact info@dspconcepts.com!