A few weeks ago I received an invite to a webinar on a Far Field Voice Dev Platform by a company called Aaware. I was curious so I attended. Basically, the Aaware far field voice development platform is a single-chip voice and sound capture solution for applications where strong interfering sources and high ambient noise are present.
It can be used to develop far-field voice interfaces for artificial intelligence (AI) applications including digital assistants, robotics, home automation and automotive. The platform can also be used to develop voice-based cloud services, including Amazon Alexa and Google Home.
The Aaware Sound Capture Platform integrates algorithmic acceleration technology and a flexible microphone array board with Avnet’s MiniZed
Zynq SoC development platform, using dual-core Xilinx
Zynq-7000 All Programmable SoC devices.
Aaware’s Acoustically Aware sound capture algorithms, paired with wake word detection technology, allow systems to adapt to differing noise interference, without requiring calibration for different environments or integration of the reference signal. The Aaware platform has a concentric circular array of 13 microphones that may be configured in different combinations, allowing for performance tuning. Its sound capture algorithms include source localization, proprietary spatial/spectral/temporal source separation, and noise suppression to deliver low distortion audio of acoustic sources in the sound field. Some of the features include:
Microphone Array Shield
- 13 MEMs microphones (MP34DB02TR)
- 15 LEDs
- Stereo audio DAC output
- Arduino-compatible shield interface
MiniZed (7010)
- Xilinx Zynq 7010 SoC
- Micron 512 MB DDR3L
- Micron 8 GB eMMC
- Murata WiFi and Bluetooth
- Arduino-compatible shield interface
Aaware reference design Aware sound capture algorithms
- Source localization and separation
- Interference cancellation of nearby sources without a reference
What do you think? How would you use it? What would you design?
You comments are appreciated.
