NeuroVoice

Ultra-low Power Tiny AI Chip for Hearables

NeuroVoice, a Tiny AI based on the Neuromorphic Analog Signal Processor, addresses the most annoying problems of mass-market hearables:  hearing experience in case of irregular noises, power consumption,  cloud connectivity and data privacy issues

Voice Extraction Provides The Best Hearing Experience!

Always-on, adaptive to different types of noise, ultra-low-power and on-device, cloud-independent voice processing

Digital hearing solutions have challenges:

1Digital sound processing is power hungry

2Digital solutions can’t provide low latency needed for real-time voice processing

3Active noise cancellation is helpless against irregular noises

Challenges answered by NeuroVoice!

1Ultra-low power consumption below 150µW

2NASP ensure ultra-fast inherence in up to 20µs

3Voice extraction delivers clear voice in any background conditions

NeuroVoice is different from other solutions in the market of affordable hearables. Typical digital devices are power hungry. But hearables require operating times of  24-48 hours on the battery charge. In contrast, NeuroVoice enables by orders of magnitude longer always-on operation due to analog neuromorphic computations performed directly on-device.  At the same time, NeuroVoice provides ultra-low latency for unsurpassed quality of hearing experience.

Digital Signal Processing vs NASP Capabilities
Solution Acoustic Voice Processing Mixed Voice Processing Irregular Noise Standard Noise Single Microphone Solution Transparent Mode AI Support
DSP
Yes
No
No
Yes
No
No
NASP
Yes
Yes
Yes
Yes
Yes
Yes

Possible applications: earbuds and smartphones, hearing aids and hearing assistance, gamers headphones and intercom systems

Additional Voice Features

  • Voice Activity Detection

    Detects the presence or absence of voice in the sound

  • Wake Word Detection

    Recognizes branded and custom wake-words in the audio stream

  • Keyword Spotting

    Receives an audio stream and sends the spotted keywords into MCU for execution

NeuroVoice supports a range of other voice processing features, in addition to Voice Extraction. Voice Activity Detection (VAD), Wake Word Detection (WWD) and Keyword Spotting (KWS) all together or in any combination allow building a voice processing product for any use case requirements.

Possible applications include Smart Home, Smart Factory,  Smart Health and other systems with voice control and management. 

Hearing Assistance Application with NeuroVoice

Refining the ability to hear the human voice in noisy environments would be a major step forward for hearing aids, hearing assistance, conferencing, two-way radios or even gaming applications.

Voice extraction by a neural network is the answer, if implemented not in the cloud but according to Tiny ML approach –  on-device. NeuroVoice, a small low-power and low-latency chip offers exactly that. 

NeuroVoice ADVANTAGES

Neuromorphic Analog Module for Voice Extraction

Longer battery life due to ultra-low power consumption

Small IC footprint

New Application Opportunities

End-user Privacy

Short time-to-market

Neuromorphic
Analog
Signal
Processing
NO FEAR of DATA TSUNAMI
Technology