Did You Know...
...that Robert Adler has the dubious distinction of being the Father of the Couch Potato? Back in 1955 Adler was employed by what was then Zenith Radio Corp., where he was charged to invent something that would allow viewers to turn down the TV volume without leaving their chairs. After a series of flops (such as a wired contraption that people tripped over), Adler hit on the idea of using sound waves. Thus the Remote Control was born...
PatentStorm News
Make the Most of PatentStorm
See this month's Top Inventors and Most Cited Patents.
Stay on top of the latest patents by subscribing to an RSS feed.
Got questions? Ask a Patent Expert!
Registered users: Manage your profile, comments and alerts.
US Patent 6615170 - Model-based voice activity detection system and method using a log-likelihood ratio and pitch
US Patent Issued on September 2, 2003
Estimated Patent Expiration Date:
March 7, 2020Estimated Expiration Date is calculated based on simple USPTO term provisions. It does not account for terminal disclaimers, term adjustments, failure to pay maintenance fees, or other factors which might affect the term of a patent.
Estimated Patent Expiration Date:
March 7, 2020Estimated Expiration Date is calculated based on simple USPTO term provisions. It does not account for terminal disclaimers, term adjustments, failure to pay maintenance fees, or other factors which might affect the term of a patent.AbstractA system and method for voice activity detection, in accordance with the invention includes the steps of inputting data including frames of speech and noise, and deciding if the frames of the input data include speech or noise by employing a log-likelihood ratio test statistic and pitch. The frames of the input data are tagged based on the log-likelihood ratio test statistic and pitch characteristics of the input data as being most likely noise or most likely speech. The tags are counted in a plurality of frames to determine if the input data is speech or noise.Other References
| InventorsApplicationNo. 09/519960 filed on 03/07/2000US Classes:704/233, Detect speech in noise704/231RecognitionField of Search704/205, Frequency704/207, Pitch704/233, Detect speech in noise704/1LINGUISTICSExaminersPrimary: To, Doris H.Assistant: Opsasnick, Michael N. Attorney, Agent or FirmUS Patent References5812965, Process and device for creating comfort noise in a digital speech transmission systemIssued on: 09/22/1998 Inventor: Massaloux6009391, Line spectral frequencies and energy features in a robust signal recognition system Issued on: 12/28/1999 Inventor: Asghar, et al.6070136, Matrix quantization with vector quantization error compensation for robust speech recognition Issued on: 05/30/2000 Inventor: Cong, et al.6219642, Quantization using frequency and mean compensated frequency input data for robust speech recognition Issued on: 04/17/2001 Inventor: Asghar, et al.6240386, Speech codec employing noise classification for noise compensation Issued on: 05/29/2001 Inventor: Thyssen, et al.6349278, Soft decision signal estimation Issued on: 02/19/2002 Inventor: Krasny, et al.6351731Adaptive filter featuring spectral gain smoothing and variable noise multiplier for noise reduction, and method therefor Issued on: 02/26/2002 Inventor: Anderson, et al. International ClassesG10L 11/02 (20060101)G10L 11/00 (20060101) |
