Solution Summary
Voice Recognition Module is a local voice trigger and commands recognition solution with robust noise reduction that consumes very low resources and provides high accuracy performance.
It runs on MCU with 120Mhz speed and supports multiple languages as well as a H/W crypto engine for security.
Features/Benefits
- Local voice recognition algorithm, no network connectivity needed
- Phoneme based modeling
– Quick command customization: don’t have to collect speech data in advance
– Optimization by model adaptation just with a small amount of speech data - Global language support: 40+ languages worldwide
- Small footprint (Flash : 200KB + 28B* Xc , RAM: 50KB+116* Xc ) and cost effective (single DMIC + MCU)
- Training Tool: command customization, performance tuning, testing
- Separation of recognition engine and command model, switching commands dynamically
- Low power and high-efficiency MCU with strong security function
Target Markets and Applications
- IoT devices
- Home appliances
- Wearable/Hearable devices
Please contact howchip@howchip.com to inquire about the purchase and further information.