Multi-microphone system to support bandpass filtering for analog-to-digital conversions at different data rates

Abstract

One exemplary problem addressed by the techniques disclosed herein is that the A-to-D converter of the broad-band audio codec that is being used more frequently in IP telephony, G.722, samples the analog audio source 16,000 times per second, rather than 8,000. Since all G.722-capable telephones must continue to be G.711-capable, one problem is that a microphone that provides appropriate bandpass filtering for G.722 encoding fails to provide adequate filtering for G.711. One exemplary aspect is therefore directed to telephones that must be able to switch back and forth between narrow-band digital audio encoding in which the A-to-D converter samples the audio stream 8,000 times per second, and a wide-band audio encoding in which the A-to-D converter samples the audio stream 16,000 times per second. This is accomplished using one or more of a plurality of switched microphones, a filter and a modification of the resonant frequencies of a handset.

Description

CROSS REFERENCE TO RELATED APPLICATION[0001]The present application is a divisional of U.S. patent application Ser. No. 12 / 882,929, filed Sep. 15, 2010, entitled “MULTI-MICROPHONE SYSTEM TO SUPPORT BANDPASS FILTERING FOR ANALOG-TO-DIGITAL CONVERSIONS AT DIFFERENT DATA RATES”, which is incorporated herein by this reference in its entirety.FIELD[0002]An exemplary aspect is directed toward communications hardware. More specifically, an exemplary aspect is directed toward enabling bandpass filtering in devices that must support more than one A-to-D conversion rate by equipping the devices with separate microphones that are physically optimized for their different rates.BACKGROUND[0003]If one wished to digitize an analog audio signal in which all of the important acoustic information is below 4,000 Hz (hertz), the basic steps of the analog-to-digital conversion would include the following:[0004]Filter from the original signal all information about 4,000 Hz,[0005]Divide the filtered signa...

AI Technical Summary

Benefits of technology

The patent describes a way to reduce the amount of computing needed for a certain task by using a technique called bandpass filtering. This filtering helps to remove certain frequencies from an input signal, which can make it easier to create a digital representation of the signal. The technical effect of this technique is that it can make the process of analyzing signals faster and more efficient, which could be useful in a variety of applications.

Problems solved by technology

A problem with speech digitization techniques that have an upper frequency limit of 4,000 Hz, such as G.711, is that many components of human speech that are important for intelligibility—for example, the acoustic information that allows “f” and “s” sounds to be distinguished from each other—are at frequencies above 4,000 Hz.

Images