AI and DSP performance leader with up to 4X the NN performance and
2X the DSP performance of the HiFi 4 DSP, ideal for digital assistants, infotainment, and voice-controlled products
Blending a neural network (NN) with digital signal processing, the Cadence® Tensilica® HiFi 5 DSP easily executes the two complementary aspects of speech recognition: audio pre-processing and speech recognition/keyword detection. Voice-controlled appliances, whether in the home or in automobiles, can achieve rich voice interactions with the HiFi 5 DSP's leading-edge performance.
Tensilica HiFi 5 DSP
Overview
Key Features
- Two 128-bit loads per cycle
- 2X MAC capability versus HiFi 4 DSP for pre- and post-processing
- 4X MAC capability versus HiFi 4 DSP for NN processing, includes 32 16x8 or 16x4 MACs per cycle
- Optional eight single-precision floating-point MACs per cycle". Update as: "8 single-precision floating-point MACs per cycle (optional)
- Optional 16 half-precision floating-point MACs per cycle". Update as: "16 half-precision floating-point MACs per cycle (optional)
- HiFi NN library offers a highly optimized set of library functions commonly used in NN processing (especially speech), which can easily be integrated into popular machine learning frameworks
- Software compatibility with the complete HiFi DSP product line, totaling over 300 HiFi-optimized audio and voice codecs and audio enhancement software packages
Applications
- Automotive,
- Communications,
- Consumer Electronics,
- Data Processing,
- Industrial and Medical,
- Military/Civil Aerospace,
- Others
Technical Specifications
Maturity
Available on request