FIELD: digital processing of speech for man-machine dialog, automatic handling of telephone networks, medicine, etc. SUBSTANCE: device has word beginning and end analyzer, word spacing segmentation unit, processing unit incorporating mean zero-crossings meter, and signal mean energy meter, as well as series-connected pre-classification unit and final classification unit whose output functions as device output; novelty is that device has, in addition, random-access memory, word length meter, analog-to-digital level converter, and analog-to-digital zero-number converter; first output of word beginning and end analyzer is connected to data input of random-access memory whose first and second outputs are connected, respectively, to data inputs of analog-to-digital level and zero-number converters and their outputs are connected, respectively, to first and second inputs of pre- classification unit whose third input is combined with clear inputs of both analog-to-digital converters and to output of word spacing segmentation unit; reference inputs of both analog-to-digital converters are connected, respectively, to outputs of signal mean energy meter and mean zero crossings meter whose inputs are connected to first output of word beginning and end analyzer whose second output is connected to combined write/read inputs of random- access memory and to input of word length meter whose output is connected to input of word spacing segmentation unit. EFFECT: improved precision of identifying isolated words irrespective of rate, volume, and main-current frequency of announcer's speech. 1 dwg
Title | Year | Author | Number |
---|---|---|---|
PROCESS OF COMPRESSION OF INSULATED LAYERS | 2000 |
|
RU2180974C2 |
ANALYZER OF RANDOM PROCESSES | 1991 |
|
RU2012052C1 |
SYSTEM AND METHOD OF SPEECH RECOGNITION | 2011 |
|
RU2466468C1 |
SPEECH RECOGNITION METHOD BASED ON TWO-LEVEL MORPHOPHONEMIC PREFIX GRAPH | 2015 |
|
RU2597498C1 |
METHOD FOR HYBRID GENERATIVE-DISCRIMINATIVE SEGMENTATION OF SPEAKERS IN AUDIO-FLOW | 2013 |
|
RU2530314C1 |
METHOD FOR RECOGNIZING SPOKEN WORDS | 2005 |
|
RU2296376C2 |
USER VERIFICATION METHOD IN AUTHORISED ACCESS SYSTEMS | 2007 |
|
RU2351023C2 |
METHOD FOR COMPRESSION OF VIDEO SIGNAL IN DIGITAL REPRESENTATION | 1990 |
|
RU2012157C1 |
METHOD OF IDENTIFYING SPEAKER FROM ARBITRARY SPEECH PHONOGRAMS BASED ON FORMANT EQUALISATION | 2009 |
|
RU2419890C1 |
METHOD AND DEVICE FOR CLASSIFYING NOISY VOICE SEGMENTS USING MULTISPECTRAL ANALYSIS | 2014 |
|
RU2606566C2 |
Authors
Dates
1999-08-27—Published
1998-01-05—Filed