QUANTISATION OF CLASSES FOR DISTRIBUTED RECOGNITION OF SPEECH Russian patent published in 2009 - IPC G01L11/04 

Abstract RU 2348019 C2

FIELD: physics, computer facilities.

SUBSTANCE: invention relates to the systems of distributed recognition of speech and describes the system, method and computer-compatible data carrier for quantisation of the information on a class and the information on the basic sound tone. The method in the data processing system comprises reception of a sound and capture of the sound frame. The method includes also the definition of the basic tone of the frame and calculation of the keyword representing the basic tone of the frame. Note here that the first meaning of the keyword specifies uncertain basic tone. The method comprises also the definition of the frame class. Note here that the class represents either of, at least two classes specifying uncertain basic tone and, at least, one class specifying certain basic tone. The method includes also the calculation of the keyword representing a class of the frame. Note that the keyword length makes maximum a minimum quantity of bits, required for representation of, at least, two classes, and a minimum quantity of bits required for representation of, at least, one class.

EFFECT: compact representation of information about class and basic tone for preservation bit transfer low speed.

19 cl, 12 dwg, 1 tbl

Similar patents RU2348019C2

Title Year Author Number
PITCH QUANTISATION FOR DISTRIBUTED SPEECH RECOGNITION 2004
  • Ramabadran Tenkasi V.
  • Sorin Aleksandr
RU2331932C2
DECISION ON PRESENCE/ABSENCE OF VOCALIZATION FOR SPEECH PROCESSING 2014
  • Gao Yan
RU2636685C2
METHOD AND DEVICE FOR EFFICIENT MASKING OF DELETED SHOTS IN SPEECH CODERS ON BASIS OF LINEAR PREDICTION 2003
  • Zhelinek Milan
  • Gurnaj Filipp
RU2325707C2
METHOD AND DEVICE FOR EFFICIENT MASKING OF DELETION OF FRAMES IN SPEECH CODECS 2006
  • Vajankur Tommi
  • Zhelinek Milan
  • Gurnaj Filipp
  • Salami Redvan
RU2419891C2
IMPROVEMENT OF CLASSIFICATION BETWEEN TIME-DOMAIN CODING AND FREQUENCY-DOMAIN CODING 2015
  • Gao Yan
RU2667382C2
PRINCIPLE FOR CODING AUDIO SIGNAL AND DECODING AUDIO SIGNAL USING INFORMATION FOR GENERATING SPEECH SPECTRUM 2014
  • Fuks Gijom
  • Multrus Markus
  • Ravelli Emmanuel
  • Shnell Markus
RU2646357C2
ADAPTIVE BAND EXTENSION AND DEVICE THEREFOR 2014
  • Gao Yan
RU2641224C2
PRINCIPLE FOR CODING AUDIO SIGNAL AND DECODING AUDIO USING DETERMINED AND NOISE-LIKE DATA 2014
  • Fuks Gijom
  • Multrus Markus
  • Ravelli Emmanuel
  • Shnell Markus
RU2644123C2
METHODS AND DEVICES OF SOURCE-GUIDED BROADBAND SPEECH CODING AT VARIABLE BIT RATE 2003
  • Zhelinek Milan
RU2331933C2
METHOD AND DEVICE FOR QUANTIZING AMPLIFICATION IN BROADBAND SPEECH ENCODING WITH ALTERNATING BITRATE 2004
  • Zhelinek Milan
  • Salami Redvan
RU2316059C2

RU 2 348 019 C2

Authors

Ramabadran Tenkasi V.

Sorin Aleksandr

Dates

2009-02-27Published

2004-02-05Filed