Noise suppression in beam-steered microphone array

• US 7,092,882 B2
• Filed: 12/06/2000
• Issued: 08/15/2006
• Est. Priority Date: 12/06/2000
• Status: Active Grant

• Current Assignee: NCR Voyix Corporation
• Original Assignee: NCR Corporation
• Inventors: Miller, Michael S., Arrowood, Jon A.
• Primary Examiner(s): CHAWAN, VIJAY B
• Application Number:
  US09/731,084
• Publication Number:
  US 20020069054A1
• Time in Patent Office: 2,078 Days
• Field of Search: 704/226, 704/233, 704/231, 704/270, 704/275
• US Class Current: 704/233
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 21/0208 Noise filtering

Abstract

A system for suppressing unwanted signals in steerable microphone arrays. The lobes of a steerable microphone array are monitored, to identify lobes having large speech content and low noise content. One of the identified lobes is then used to deliver speech to a speech recognition system, as at a self-service kiosk.

Claims

1. Apparatus comprising:

a) a self-service kiosk which dispenses articles, currency, or communication services; and

b) within the kiosk,

i) a steerable beam microphone array, having multiple lobes;

ii) means for sampling lobes, and

A) distinguishing the difference between speech content and noise content from sound signals received by each lobe,

B) identifying lobes having a relatively high speech content,

C) identifying lobes having a relatively low noise content, and

D) actuating a lobe having both a relatively high speech content and relatively low noise content.

2. Apparatus according to claim 1, and further comprising:

c) speech recognition means for recognizing speech contained in the lobe actuated.

3. A method, comprising the following steps:

a) maintaining a self-service kiosk which dispenses articles, currency, or communication services;

b) maintaining a beam-steerable microphone array at the self-service kiosk;

c) measuring noise content and speech content of several lobes of the array; and

d) selecting a lobe which carries

i) larger speech signals than other lobes and

ii) smaller noise signals than other lobes.

4. Method according to claim 3, and further comprising the step of:

e) receiving signals from the lobe selected, and performing speech recognition on the data.