Processing

Please wait...

Settings

Settings

1. WO2000042600 - METHOD IN SPEECH RECOGNITION AND A SPEECH RECOGNITION DEVICE

Publication Number WO/2000/042600
Publication Date 20.07.2000
International Application No. PCT/FI2000/000028
International Filing Date 17.01.2000
Chapter 2 Demand Filed 14.08.2000
IPC
G06F 3/16 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
3Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
16Sound input; Sound output
G10L 25/87 2013.01
GPHYSICS
10MUSICAL INSTRUMENTS; ACOUSTICS
LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
25Speech or voice analysis techniques not restricted to a single one of groups G10L15/-G10L21/129
78Detection of presence or absence of voice signals
87Detection of discrete points within a voice signal
CPC
G10L 25/87
GPHYSICS
10MUSICAL INSTRUMENTS; ACOUSTICS
LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
25Speech or voice analysis techniques not restricted to a single one of groups G10L15/00-G10L21/00
78Detection of presence or absence of voice signals
87Detection of discrete points within a voice signal
Applicants
  • NOKIA MOBILE PHONES LTD [FI/FI]; Keilalahdentie 4 FIN-02150 Espoo, FI (AllExceptUS)
  • LAURILA, Kari [FI/FI]; FI (UsOnly)
  • HÄKKINEN, Juha [FI/FI]; FI (UsOnly)
  • HARIHARAN, Ramalingam [IN/FI]; FI (UsOnly)
Inventors
  • LAURILA, Kari; FI
  • HÄKKINEN, Juha; FI
  • HARIHARAN, Ramalingam; FI
Agents
  • TAMPEREEN PATENTTITOIMISTO OY; Hermiankatu 6 FIN-33720 Tampere, FI
Priority Data
99007818.01.1999FI
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) METHOD IN SPEECH RECOGNITION AND A SPEECH RECOGNITION DEVICE
(FR) PROCEDE ET DISPOSITIF DE RECONNAISSANCE DE LA PAROLE
Abstract
(EN)
In a method for detecting pauses in speech in speech recognition, for recognizing speech commands uttered by the user, the voice is converted into an electrical signal, whose frequency spectrum is divided into two or more sub-bands. Samples of the signals on the sub-bands are stored at intervals, the energy levels of the sub-bands are determined on the basis of the stored samples, a power threshold value (thr) is determined, and the energy levels of the sub-bands are compared with said power threshold value (thr). The comparison results are used for producing a pause detecting result.
(FR)
L'invention concerne un procédé permettant de détecter les pauses dans un discours, dans la reconnaissance de la parole, ainsi que les ordres énoncés par l'utilisateur, dans lequel la voix est transformée en un signal électrique dont le spectre de fréquence est divisé en au moins deux sous-bandes. Le procédé consiste à stocker des échantillons des signaux des sous-bandes à intervalles donnés, à déterminer les niveaux énergétiques des sous-bandes d'après les échantillons stockés, à déterminer une valeur seuil de puissance (thr) et à comparer les niveaux énergétiques des sous-bandes avec la valeur seuil (thr). Les résultats de cette comparaison sont utilisés pour produire un résultat de détection de pause.
Also published as
Other related publications
Latest bibliographic data on file with the International Bureau