SPEAKER DIALIZATION CORRECTION METHOD AND SYSTEM UTILIZING TEXT-BASED SPEAKER CHANGE DETECTION

To provide a speaker dialization correction method and system utilizing text-based speaker change detection.SOLUTION: A speaker dialization correction method includes the steps of: executing speaker dialization on an input audio stream; recognizing a voice included in the input audio stream to be co...

Full description

Saved in:
Bibliographic Details
Main Authors HEO HEE SOO, KWON YOUNGKI, LEE BONG JIN, JUNG NAMKYU, KIM GEONMIN, LEE CHAN KYU
Format Patent
LanguageEnglish
Japanese
Published 19.08.2022
Subjects
ACOUSTICS
MUSICAL INSTRUMENTS
PHYSICS
SPEECH ANALYSIS OR SYNTHESIS
SPEECH OR AUDIO CODING OR DECODING
SPEECH OR VOICE PROCESSING
SPEECH RECOGNITION
Online AccessGet full text

Cover

More Information
Summary:To provide a speaker dialization correction method and system utilizing text-based speaker change detection.SOLUTION: A speaker dialization correction method includes the steps of: executing speaker dialization on an input audio stream; recognizing a voice included in the input audio stream to be converted into a text; detecting speaker change based upon the converted text; and correcting the speaker dialization based upon the detected speaker change.SELECTED DRAWING: Figure 3 【課題】テキストベースの話者変更検出を活用した話者ダイアライゼーション補正方法及びシステムを提供する。【解決手段】話者ダイアライゼーション補正方法は、入力されたオーディオストリームに対する話者ダイアライゼーションを実行する段階と、入力されたオーディオストリームに含まれた音声を認識してテキストに変換する段階と、変換されたテキストに基づいて話者変更を検出する段階と、検出された話者変更に基づいて前記話者ダイアライゼーションを補正する段階と、を含む。【選択図】図３
Bibliography:Application Number: JP20220012297