DSP Concepts announces support for Cadence Tensilica HiFi 5 DSP in its Audio Weaver platform with a public release planned for the end of the second quarter of 2021. HiFi 5 is used for low-power wireless products, headphones, wearable devices, and high-end automotive systems. A full suite of DSP concepts for audio processing and IP operation will be made available to allow OEMs to create differentiated products that take advantage of Audio Weaver technology. In an interview with EEWeb, Paul Beckmann, Chief Technology Officer and founder of DSP Concepts, highlighted how Tensilica HiFi 5 DSP includes hardware acceleration to enable voice-controlled user interfaces and artificial intelligence (AI)-based voice applications. “The complexity of audio products continues to increase especially as more product makers want to add voice and AI features,” Beckmann said. He added, “The HiFi-5 processor is ideal for these applications and has been designed with high computing throughput and low power consumption. The HiFi-5 will enable the next generation of smart wearables with context awareness, a built-in voice assistant, and long battery life.” Audio Processing Digital signal processing, whatever its nature, is undoubtedly one of the technologies that has received the most success and attention in recent years, with multiple applications in various fields of electronics and applied sciences in general. Voice information is simultaneously playing an increasingly important role in the context of both industrial and consumer applications, with the development of different technologies. Poor audio processing kills the productivity of audio conferencing. Hearing details without unwanted noise, echo, or distortion, especially in conference calls, is critical and involves many techniques. “Every audio product is different and you have to think carefully about the algorithms required including microphone input for audio playback and loudspeaker for output. If the product has an amplifier, you will need an Acoustic Echo Cancellation (AEC) device. If it has multiple speakers, such as an amplifier For audio, then a multi-channel AEC should be needed. “Array microphones improve product performance in noisy environments and with echo,” Beckman said, “Just like using Tensorflow for machine learning development, Audio Weaver is used to develop audio products.” “Audio Weaver contains all the parts with a lower level of IP (beam converters, echo cancellation modules, noise reduction, speaker equalization, etc.) and product makers can use Audio Weaver to design their products quickly and efficiently.” Digital signal processors are a computer architecture to implement Audio/Visual operations that inherently refer to signal information and therefore require extensive computational processing activities.Field programmable gate arrays and MCUs represent another computational solution thanks to the possibility of implementing processor algorithms Audio/visual signals using programmable devices. The alternative between these solutions is based on a common requirement for audio/visual applications, and real-time processing. The real-time requirements are particularly stringent for audio (but also visual) applications because the computational level is particularly high, especially for the visual component. “MCUs, DSPs, and application processors can all be used to process audio. Real time means that the audio workload must be able to complete within a specified amount of time. This requires careful identification of the processor to ensure that the workload is appropriate. This requires accurate classification tools that work in real time and can identify hot spots where significant processing is occurring,” Beckman said. He added, “DSPs have advantages over MCUs and application processors when it comes to power consumption. The DSP architecture is optimized for mathematical operations and has a high degree of parallelism. We have found that the HiFi-5 DSP requires the fewest number of instructions to complete commonly used DSP functions such as filters and echo cancellation devices. A more efficient processor architecture translates into lower power consumption.” The main challenge for DSPs is the difficulty of programming and optimizing them. However, with Audio Weaver, Beckmann explained that DSP concepts have dealt with low-level programming and optimization allowing engineers to graphically design and fine-tune algorithms. Figure 1 : DSP Cadence Tensilica HiFi 5 DSP Hardware and Software The exponential growth in the adoption of voice interactions for a wide variety of devices has led to the need for increased signal processing and neural network performance.To do this, Cadence is working with DSP Concepts to enable processing algorithms on the Tensilica HiFi 5 DSP to improve bidirectional communication and performance of voice user interfaces and processing operation.Cadence notes that this fifth generation HiFi DSP delivers four times the processing performance of the HiFi 4 DSP, making it an ideal solution for voice-controlled user interfaces in digital home assistants and in-vehicle infotainment system. 5 DSP Enhanced fixed and floating point DSP capabilities and native support for new data types, saving memory and power. Lower latency, increased privacy, and more natural voice user interface interactions that consumers demand, the processing load on the device is rapidly increasing. HiFi 5 DSP with Audio Weaver delivers the performance needed to handle front-end processing tasks, including echo cancellation and noise reduction. At the heart of these software and hardware innovations are smart turnkey solutions that include a far-reaching voice user interface for smart assistants, TVs, speakers, and receivers. These solutions are also optimized to improve home and office audio and video conferencing, supporting 2-mic, 4-mic, 6-microphone and custom configurations with AEC for up to 7 channels. “Both wearable devices and conferencing devices can take advantage of AI for voice. In both applications, AI will be used to eliminate noise and focus on the main speaker. AI can be used to authenticate the owner of the wearable device. The most advanced AI algorithms can be used in meeting room application because it has a larger processor and no battery limitations.Cadence’s upcoming HiFi 5 processor will allow complex AI applications even at low power.This will expand AI applications in wearable devices, especially wireless systems that are very energy sensitive. “. In combination with enhanced Audio Weaver libraries, Cadence’s Tensilica HiFi 5 DSP will have access to DSP Concepts’ full suite of Audio Front End (AFE) offering multi-channel echo cancellation solutions with advanced beamforming using 8 or more microphones. Additionally, new machine learning (ML) instruction sets in the Tensilica HiFi 5 DSP improve full duplex communication and actuation processing. Audio Weaver aims to provide clear audio communication in the most challenging environments. Beckman noted that the library not only provides the processing required for a correct audio user interface but also supports the power performance of all battery-powered solutions. Audio Weaver allows designers to create complex digital audio applications without writing a single line of code. Developers simply select the necessary audio modules from a library of more than 500, graphically link them, and adapt them to the target devices in an equally simple process; The built-in examples speed up learning and allow new designs to be implemented more quickly. The speaker highlighted that DSP Concepts has a set of rapid prototyping devices that can be used early on for development. These devices can be used early on to answer performance questions and product development to eliminate risks. This article belongs to the EE Times. Please visit this link for the full article. .