Microsemi has audio DSP platform for intelligent processing
Microsemi's new Timberwolf platform is described as a third generation, low power DSP architecture with a proprietary 4 MAC core, dedicated hardware accelerators and two independent 16bit D/A converters in a single 64pin QFN and 56pin 3x3 CSP packages.
When combined with algorithmic firmware, Timberwolf can enable beam forming, multi-channel acoustic echo cancellation, direction of arrival and far field MIC capabilities.
It is aimed at hands-free applications that require clear speech, free of noise and echo.
"These new advanced DSP technologies are the foundation for improved sound quality, automatic speech recognition, sound classification and other intelligent decision making functions based on sound and intelligent processing," said Microsemi in a statement.
Hardware features include a 300MHz proprietary DSP architecture, standalone auto-boot from flash, two clock independent TDM ports, 14 GPIOs, two digital microphone interfaces, dual 16bit D/A converters and headphone amps capable of four single-ended or two differential outputs.
