News | June 30, 1998

SRS Develops DSP Code for VIP Technology

SRS Labs Inc. announced the availability of its Voice Intelligibility Processor (VIP) technology in digital code that can be ported to any of the industry-standard Digital Signal Processors (DSPs). Toshiba Corporation is the first of SRS Labs' semiconductor partners to port the VIP code into its TC9447F-006 chip.

SRS Labs acquired the VIP technology in January 1998. VIP is a patented voice technology that significantly improves the intelligibility of the human voice across a range of listening situations, particularly in high ambient noise environments. VIP has been successfully implemented in projects for NASA, the U.S. Post Office, the Coca-Cola Syrup Plant and Shea Stadium, among others.

Intelligibility of human voice is derived largely from the pattern of the frequency distribution of voice formants. In acoustic phonetics, formants collectively constitute the frequency spectrum of a speech sound. The relative positioning of the first and second formants is usually sufficient to distinguish a sound from all others. Formants are somewhat comparable to harmonics on a musical instrument, more accurately to the partials of a tone. Therefore, selective enhancement of the formants provides greater intelligibility of speech especially in the presence of high background noise with minimal increase in the overall signal peak level.

The VIP process functions in real-time using digital signal analysis and processing. During the VIP process, formants are processed by individually selecting them in a real-time spectrum analyzer and individually amplifying and selectively weighting them before recombining processed formants and unprocessed base band voice components to provide an output signal of greatly improved intelligibility.

The VIP technology can be used in cellular phones, wireless pagers, telecom handsets, speakerphones, and hearing aids.

Toshiba's TC9447F-006 is self-contained, requiring no external processors to operate. The TC9447F-006 contains built in analog to digital, and digital to analog converters. It operates with two channels at a 32kHz-sampling rate.