Key Takeaways:
- Voice AI using Small Language Models is becoming a major market requirement for phones, appliances, and immersive audio.
- Cadence Design is a leading supplier of Intellectual Property (IP) for custom audio processing chips.
- The new Tensilica HiFi iQ platform enables the use of Small Language Models at higher performance and lower power consumption.
- The audio processing market is large, with 1.5 billion DSPs shipping annually with Cadence logic.
- The innovation in audio processing will continue, with applications such as spatial audio, language translation, and personalized sound experiences.
Introduction to Voice AI and Audio Processing
Voice AI using Small Language Models is becoming a major market requirement for various devices, including phones, appliances, and immersive audio systems. This technology is revolutionizing the way we interact with devices, making them more intuitive and user-friendly. Many analysts believe that AI will become pervasive, applied to every aspect of computing to improve product quality, increase productivity, and lower costs. While some investors are cautious about the potential AI bubble, most analysts see meaningful adoption of AI technology in various industries. One area that is often overlooked is the processing of audio content, which has seen significant improvements in recent years.
Cadence Design and Audio Processing
Cadence Design is a leading supplier of Intellectual Property (IP) for custom audio processing chips. The company’s Tensilica team has been at the forefront of audio processing technology, and their IP is used in a wide range of devices, from headphones to TVs and automobiles. The Tensilica team has recently revamped their IP to enable the use of Small Language Models at higher performance and lower power consumption. The new platform, called the Tensilica HiFi iQ, is a sixth-generation DSP for audio and voice processing. This platform is designed to provide higher quality and dynamic range, making it suitable for high-end applications.
The Tensilica HiFi iQ Platform
The Tensilica HiFi iQ platform is a fully customizable platform that customers can tailor to their specific product needs. The platform features low-precision math, better auto-vectorization, and raw performance enabled by a 256-bit broad vector instruction set. Additionally, the platform includes AI performance enhancements, such as multiply-accumulate cores, which are necessary for matrix multiplication. The new platform will be widely available in the second quarter of 2026, allowing design teams to produce customized solutions for their specific product needs. The Tensilica HiFi iQ platform is a significant improvement over previous generations, providing higher performance and lower power consumption.
The Automobile: A Great Example
The automobile industry is a great example of the application of audio processing technology. Modern cars have a plethora of audio processing chips that enable infotainment and hands-free speech-to-intent across various applications. The higher-end of the market includes spatial 3D audio, voice agents, noise cancellation, personalization, and sound zoning. The next time you ride in a well-appointed vehicle and enjoy an immersive sound experience, it might just be brought to you by a chip with Cadence IP. The automotive market is already a major adopter of audio silicon and will continue to drive innovation and revenue.
Market Size and Growth
The market for audio processing chips is significant, with 1.5 billion DSPs shipping annually with Cadence logic. The company has over 160 licensees, and its IP is used in a wide range of devices. The market is expected to continue growing, driven by the increasing demand for high-quality audio experiences in various applications. The innovation in audio processing will be continual, with applications such as spatial audio, language translation, and personalized sound experiences becoming more prevalent.
What’s Next?
The future of audio processing is exciting, with many new applications and innovations on the horizon. Earbuds that can translate different languages and create a surround-sound experience with spatial audio are just a few examples of what’s possible. Cars will be better able to understand commands and initiate actions, while targeting each occupant with detailed and custom music and information. Audio equipment makers will lead this innovation, using Cadence’s IP in their custom ASICs to improve functionality and enable superior user experiences. The world around us is changing due to the application of AI in just about everything, and in the audio space, it’s all good!
