The new miniDSP UMA-8 is a high-performance yet low-cost multichannel USB microphone array built around XMOS multicore technology, designed for voice-recognition application development. Seven high-performance MEMS microphones are configured in a circular arrangement to provide high-quality voice capture for a wide range of applications. Leveraging the onboard DSP processing, the UMA-8 supports voice algorithms including beamforming, noise reduction, acoustic echo cancellation, and de-reverb.
Developed with applications in voice-activated control, smart assistants, robotics, conferencing, and more in mind, the UMA-8 pocket-size platform targets both DIYers and the OEM market, and was engineered for flexibility in firmware, software and hardware. For advanced users, full control and configuration of the DSP array processing parameters are available with a real-time GUI. This can be used to fine tune the various algorithms: acoustic echo cancellation, noise reduction, voice activation detect, and so on, dramatically improving voice pickup.
The UMA-8 costs $95. Step-by-step application notes are available for setup and configuration of the UMA-8 with the most common smart assistants currently available, including Amazon Alexa Voice + Raspberry Pi, Microsoft Cortana, and Apple Siri. miniDSP will be expanding those application notes in the future.
miniDSP | www.minidsp.comSponsor this Article
Circuit Cellar's editorial team comprises professional engineers, technical editors, and digital media specialists. You can reach the Editorial Department at email@example.com, @circuitcellar, and facebook.com/circuitcellar