DSP Group’s DBMD5/DBMD5P audio/voice systems-on-chip (SoCs) enable far-field voice user interfaces and voice communications for smart devices. Both SoCs are equipped with superior far-field beamforming and acoustic echo cancellation (AEC) technologies and support multiple microphones while enabling a wide range of capabilities. The SoCs’ pre-processing front-end algorithms are optimized to ensure high performance of local, embedded and remote cloud automated speech recognition (ASR) engines under difficult conditions.
The DBMD5/DBMD5P support both single and multiple embedded ASR engines to enable wake-word (WW) detection, seamless query recognition, and optional local voice commands. Meeting all voice requirements including full-duplex voice calls, the power-efficient, comprehensive solutions include a software pack, debug utilities, tuning tools, and documentation.