DSP Concepts' Audio Weaver Platform Supports Cadence Tensilica HiFi 5 DSP

Article by: Maurizio Di Paolo Emilio

DSP Concepts announces the support of the Cadence Tensilica HiFi 5 DSP in its Audio Weaver platform. A general version is planned for the end of the second quarter of 2021.

DSP Concepts announces the support of the Cadence Tensilica HiFi 5 DSP in its Audio Weaver platform. A general version is planned for the end of the second quarter of 2021. The HiFi 5 is used for low power wireless products, headsets, wearables, and high-end automotive systems. DSP Concepts offers a full suite of voice processing and playback IP addresses that OEMs can use to create differentiated products using Audio Weaver technology.

In an interview with EE Times, Paul Beckmann, CTO and founder of DSP Concepts, highlighted how the Tensilica HiFi 5 DSP includes hardware acceleration to enable voice-controlled user interfaces and audio applications based on artificial intelligence (AI).

"The complexity of audio products continues to increase, especially as more and more product manufacturers want to add voice and AI functions," said Beckmann. He added: “The HiFi 5 processor is ideal for these applications and is designed with high throughput and low power consumption in mind. The HiFi-5 enables the next generation of intelligent wearables with context awareness, integrated voice assistant and long battery life. "

Audio processing
Digital signal processing, whatever its nature, is undoubtedly one of the technologies that has received the greatest success and attention in recent years, with diverse applications in various fields of electronics and applied science in general. At the same time, audio information is playing an increasingly important role in the context of industrial and consumer applications, with various technologies being developed. Poor audio processing is the loss of productivity in audio conferencing. Hearing details without unwanted noise, echoes, or distortion, especially in group meetings, is vital and involves many technologies.

“Every audio product is different and you have to carefully consider which algorithms are needed, including microphone input for voice and speaker for output. If the product has a speaker, you will need an acoustic echo canceller (AEC). If it has multiple speakers, such as B. a soundbar, a multi-channel AEC is required. A microphone array improves the product's performance in noisy and reverberant environments, ”said Beckmann.

"Just like Tensorflow is used for machine learning development, Audio Weaver is used for audio product development," Beckmann said. "Audio Weaver includes all of the IP sub-components (beamformers, echo cancellers, noise cancellation, speaker compensation, etc.), and product manufacturers can use Audio Weaver to develop their products quickly and efficiently."

Digital signal processors provide a computer architecture for implementing audiovisual processes that are inherently related to signal information and therefore require intensive computational processing activities.

Field Programmable Gate Arrays (FPGAs) and MCUs are another computing solution because audio / visual signal processing algorithms can be implemented with programmable hardware. The alternative between these solutions depends on a common requirement of audio / video applications, real-time processing. The real-time requirement is particularly high for audio applications (but also for visual applications), since the computing effort is particularly high for the visual component in particular.

“MCUs, DSPs, and application processors can all be used for audio processing. Real time means that the audio workload must be completed within a specified time interval. This requires careful profiling of the processor to ensure that the workload fits. This requires precise profiling tools that work in real time and can locate hotspots where significant processing is taking place, ”said Beckmann.

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 found that the HiFi 5 DSP required the fewest number of instructions to perform commonly used DSP functions such as filters and echo cancellers. A more efficient processor architecture leads to lower power consumption. "

The biggest challenge for DSPs is that they are more difficult to program and optimize. With Audio Weaver, Beckmann emphasized that DSP Concepts took over the simple programming and optimization so that engineers can graphically design and optimize algorithms.

Figure 1: DSP Cadence Tensilica HiFi 5. Click to enlarge the image.

DSP hardware and software
The exponential growth in the introduction of voice interactions for a variety of devices has fueled the need for increased signal processing and neural network performance. To this end, Cadence is working with DSP Concepts to enable processing algorithms on the Tensilica HiFi 5 DSP to improve full-duplex communication, voice UI performance and playback processing (Figures 1 and 2).

Figure 2: Audio Weaver Designer

Cadence points out that this fifth generation HiFi DSP offers four times the processing power of the HiFi 4 DSP, making it an ideal solution for voice-controlled user interfaces in digital home assistants and infotainment for automobiles. The HiFi 5 DSP offers improved fixed and floating point DSP functions as well as native support for new data types, saving memory and power.

The processing load on the device is increasing rapidly with the demands for lower latency, greater privacy, and more natural voice user interface interactions demanded by consumers. The HiFi 5 DSP with Audio Weaver provides the performance required for the front-end processing tasks, including echo cancellation and noise suppression.

At the heart of these software and hardware innovations are intelligent, turnkey solutions that include a far-field voice interface for intelligent assistants, televisions, sound bars and set-top boxes. These solutions are also optimized for enhanced voice communication and video conferencing at home and in the office and support 2-microphone, 4-microphone, 6-microphone and custom configurations with acoustic echo cancellation (AEC) for up to 7 channels.

“Both portable devices and conference devices can use AI for audio. In both applications, AI would be used to eliminate noise and focus on the main speaker. AI can be used to authenticate the owner of the portable device. More sophisticated AI algorithms can be used in the conference room application because they have a larger processor and there are no battery limitations. The upcoming HiFi 5 processor from Cadence enables complex AI applications even with low power consumption. This will expand AI applications in wearable devices, especially in truly wireless systems that are very performance sensitive, ”said Beckmann.

In combination with optimized Audio Weaver libraries, the Tensilica HiFi 5 DSP from Cadence can access the entire suite of Audio Front-End (AFE) TalkTo from DSP Concepts, which offers multi-channel echo cancellation solutions with advanced beam shaping with 8 or more microphones. In addition, the new machine learning (ML) instruction sets in the Tensilica HiFi 5 DSP improve full-duplex communication and playback processing.

Audio Weaver aims to provide clear voice communication in the most difficult environments. Beckmann pointed out that in addition to providing the processing necessary for a correct voice user interface, the library also supports powering all battery powered solutions.

With Audio Weaver, designers can create complex digital audio applications without writing a single line of code. Developers simply select the required audio modules from a library of more than 500 modules, link them graphically and adapt them to the target hardware in an equally simple process. The included examples accelerate learning and enable faster implementation of new designs.

The speaker highlighted that DSP Concepts has a range of rapid prototyping hardware that can be used early on for development. This hardware can be used early on to answer performance questions and reduce product development.

This article was originally published in the EE Times.

Maurizio Di Paolo Emilio has a Ph.D. in physics and is a telecommunications engineer and journalist. He has worked on various international projects in the field of gravitational wave research. He works with research institutions to develop data acquisition and control systems for space applications. He is the author of several books published by Springer as well as numerous scientific and technical publications on the subject of electronic design.

Related posts:

continue reading https://dailytechnonewsllc.com/dsp-concepts-audio-weaver-platform-supports-cadence-tensilica-hifi-5-dsp/?feed_id=6210&_unique_id=61d217a399f78