Speech detection device for detecting a speech signal in an input signal

Number of patents in Portfolio can not be more than 2000

United States of America Patent

PATENT NO 6654450
SERIAL NO

09534214

Stats

ATTORNEY / AGENT: (SPONSORED)

Importance

Loading Importance Indicators... loading....

Abstract

See full text

In a speech detection device (13) for detecting a speech signal (SS) in an input signal (ES), including input means (14) arranged to receive an input signal (ES) which can contain a speech signal (SS) and/or a tone signal (CNG, BSY), and including processing means (15) for processing the input signal (ES) and for supplying at least first and second characteristic information (FS1, FS2, FS3, FS4) of the input signal (ES), and including threshold test means (16) for checking whether the first and the second characteristic information (FS1, FS2, FS3, FS4) exceed a threshold value and for supplying at least one item of detection information (D1, D2, D3, D4), and including evaluation means (17) for evaluating the at least one item of detection information (D1, D2, D3, D4) and for supplying speech detection information (SDI) which is characteristic of the presence of a speech signal (SS) in the input signal (ES), the processing means (15) further include filter means (22) adapted to extract a first frequency band signal (FS1) from a first frequency band (FB1) of the input signal (ES) as first characteristic information and to extract a second frequency band signal (FS2) from a second frequency band (FB2) of the input signal (ES) as second characteristic information, the first frequency band (FB1) and the second frequency band (FB2) not overlapping one another, and the threshold test means (16) are adapted to supply first detection information (D1) when the first frequency band signal (FS1) exceeds a first amplitude threshold value (SW), and to supply second detection information (D2) when the second frequency band signal (FS2) exceeds a second amplitude threshold value (SW), and the evaluation means (17) are adapted to supply speech detection information (SDI) when the first detection information (D1(PZ)) and the second detection information (D2(PZ)) within a test time interval PZ appear at least for a threshold time interval (SZ).

Loading the Abstract Image... loading....

First Claim

See full text

Family

Loading Family data... loading....

Patent Owner(s)

Patent OwnerAddress
SAGEM COMMUNICATIONLE PONANT DE PARIS 27 RUE LEBLANC PARIS 75015

International Classification(s)

  • [Classification Symbol]
  • [Patents Count]

Inventor(s)

Inventor Name Address # of filed Patents Total Citations
Aigner, Othmar Modling, AT 1 1
Fleischer, Christian Vienna, AT 19 528

Cited Art Landscape

Load Citation

Patent Citation Ranking

Forward Cite Landscape

Load Citation