CEVA announced that the CEVA-BX2™ audio DSP supports Dolby MS12 multi-stream decoder.

As smart TVs, over-the-top (OTT) and set-top boxes develop into multi-functional digital media receivers, multiple content sources must be obtained using multiple audio codecs. Dolby MS12 is a comprehensive, high-efficiency, low-cost solution that reduces the complexity of integrating multiple audio technologies into these devices. Dolby MS12 supports the decoding of various high-quality audio content, including Dolby Atmos used by many content service providers such as Netflix, and Dolby AC-4, which is one of the next-generation audio standards in ATSC 3.0.

The certification of the Dolby MS12 implemented and optimized by CEVA on the CEVA-BX2 DSP will enable system-on-chip (SoC) designers and equipment manufacturers to integrate Dolby technology (including Dolby Atmos and Dolby AC-4) into their products. The powerful CEVA-BX2 DSP can also be used to run CEVA's other value-added software, such as ClearVox™ noise reduction and WhisPro™ voice recognition software support multi-microphone voice user interface.

Moshe Sheier, Vice President of CEVA Marketing, said: “CEVA has been at the forefront of audio/voice innovation and has enabled more than 7 billion audio/voice devices to date. By adding Dolby MS12 to the optimized and certified CEVA DSP A wide range of software packages will be able to promote OEM manufacturers and semiconductor customers to accelerate product development to support Dolby audio technology. We are very pleased to announce this milestone and look forward to helping our mutual customers bring smart and connected multimedia products to the market and fully Take advantage of the excellent audio experience we have achieved together."

CEVA-BX2 is a modern high-performance audio/voice DSP with AI capabilities, designed for intensive audio applications, such as multi-channel audio decoding, far-field noise reduction, and artificial intelligence-based speech recognition and sound analysis. CEVA-BX2 targets high-performance audio devices, such as smart TVs and OTT/STB media devices, smart speakers, sound bars and car infotainment systems. CEVA-BX2 uses four-channel 32X32-bit MAC and eight-channel 16X16-bit MAC, and has enhanced functions to support 16×8-bit and 8×8-bit MAC operations. Its 11-stage pipeline and 5-channel VLIW micro-architecture provide parallel processing support with dual scalar computing engines, load/store and process control. It is produced on TSMC’s 7nm process and reaches a rate of 2 GHz. The CEVA-BX2 instruction set architecture (ISA) combines single instruction multiple data (SIMD) support and an optional floating-point unit for high-precision arithmetic. CEVA-BX2 comes with a complete software development tool chain, including advanced LLVM compiler, Eclipse-based debugger, DSP and neural network computing software library, neural network framework support and real-time operating system (RTOS).