Tensilica's HiFi Audio DSP Cores

Products » Audio DSP

HiFi - The Most Popular Audio DSP Cores

Today's audio requirements range from low-power MP3 playback to high-performance home-theater multi-channel playback of recorded media. Tensilica's HiFi audio cores cover the complete range. That's why so many top-tier semiconductor companies have picked Tensilica's HiFi Audio DSPs for their products. Tensilica's HiFi DSP cores feature:

Earbuds

The lowest power (longest battery life) MP3 playback in the industry, requiring just 5.7 MHz when running at 128 Kbps, 44.1 KHz and dissipating only 0.45 mW in TSMC's 65nm LP process (including memories). This helps you achieve over 100 hours of battery life or more.

Blu-ray

The latest Blu-ray Disc codecs, including Dolby Digital Plus, Dolby TrueHD, DTS-HD High Resolution Audio and DTS-HD Master Audio. The HiFi DSPs offer significant cost savings and a simplified programming model, as all Blu-ray audio processing can be done in one core, unlike other solutions that require two or more cores.

JVC car HD

All of the codecs necessary for digital radio, anywhere in the world. The HiFi DSPs enable designers to build one chip that will work with any of the digital radio standards.
Android

Tensilica is working with MIPS to help companies speed the design of new home entertainment and mobile consumer products based on Android. The two companies have developed a demo that includes a MIPS32 processor core teamed with Tensilica's HiFi 2 Audio DSP core. Tensilica's HiFi 2 Audio DSP is an ideal audio processor because it is proven in millions of consumer devices, has low power consumption, and comes with a comprehensive library of over 60 audio codecs for just about any standard from simple MP3 to Dolby and DTS, plus a full suite of partner solutions for audio post processing. See press release.

Tensilica's HiFi Audio DSP is the most popular audio DSP core on the market today, used by industry giants including AMD, Intel, Fujitsu, Samsung and more. Why? Because we've optimized the processor to efficiently handle audio streams with specialized instructions just for audio processing tasks. And because we have the most comprehensive library of audio codecs in the industry.

Tensilica's processors and DSPs can be used for:

  1. Host operating system and control using Xtensa customizable or Diamond Standard processor cores
  2. Audio codec processing and control using Tensilica's turn-key 330HiFi Audio DSP core
  3. Audio codec processing and control plus customizations for your application - just add the HiFi 2 or HiFi EP Audio DSPs to your Xtensa LX3 processor design - they're click-box options.
  4. Audio processing your own way - design your own audio processor using Tensilica's Xtensa processor technology

Solution 1: Host OS Control

Tensilica's Diamond Standard processors and Xtensa customizable processors are the most code-efficient 32-bit processors on the market today. While they excel in the data plane, they are perfectly matched for standard control tasks, too.

Solution 2: The 330HiFi Audio DSP Core

Tensilica's 330HiFi Audio DSP Coreprovides a turn-key, fully optimized approach to embedding multiple audio standards into SOC designs. It is optimized with the HiFi 2 Audio DSP to run all of the codecs written in C in a pre-packaged, drop-in format with all software components pre-ported. You can pick from our library of over 60 pre-ported software codecs to get the software you need, available immediately. And because it's a hybrid control CPU plus audio DSP with plenty of processing headroom, in many applications you can also use it as the system control processor. Find out more about the 330HiFi Audio Processor.

Solution 3: Use the HiFi 2 or HiFi EP Click-box Option with an Xtensa Processor

If you want to use a drop-in audio DSP, but want to further optimize the core, this is the option for you. The HiFi 2 and HiFi EP Audio DPSs are click-box options for the Xtensa LX3 processor. Can't decide between HiFi 2 and HiFi EP? Find out more here.

Solution 4: Design Your Own Audio DSP

If you have highly specialized audio processing needs, you can use an Xtensa processor and design your own audio DSP. By adding your own audio-specific instructions and customizing the datapath for the audio stream, you can design your own specialized audio solution, or you can start with the HiFi 2 Audio Engine and add additional specialized pre-processing or post-processing accelerator instructions.

HiFi 2 or HiFi EP?

HiFi EP is a superset of HiFi 2 with advanced optimizations for DTS Master Audio, improved voice pre- & post-processing, and improved cache memory subsystem.

  • HiFi EP includes a novel and unique 32x24 MAC for higher performance at lower power on the popular DTS Master Audio lossless decoder, resulting in a clock rate reduction of almost 35% compared to HiFi 2.
  • To address the increasingly demanding requirements in mobile and VoIP applications for better immunity to background noise and speakerphone mode quality, new instructions have been added to accelerate voice pre- and post-processing for noise cancellation and beam forming microphones.   These instructions also provide better general DSP capabilities.
  • The cache memory subsystem is enhanced with an integrated predictive prefetch unit to significantly improve performance in SOC designs with large external memory latencies.

Audio and Voice Software Packages

The following audio packages are available for the HiFi 2 Audio Engine:

The following voice packages are available for the HiFi 2 Audio Engine:


Because the Xtensa HiFi Audio DSPs are programmable, multiple standards can run on the same hardware, allowing the same device to play or record digital audio in different standards. They also can be used for many control functions since they are built on a 32-bit Xtensa RISC processor. Designers can further customize the underlying Xtensa LX processor so it can be used for other functions on the chip.

Simplified Programming

A key advantage of the HiFi Audio DSPs is the simple Xtensa processor-based programming model used in both DSPs. Because of the efficiency of the optimized audio and voice instructions, software developers can port audio and voice codecs completely in C while maintaining or surpassing the performance of assembly on other DSP and CPU architectures.

Marketing Agency