Back to results
US 10,475,441 B2Granted
Voice end-point detection device, system and method
AssigneesHyundai Motor Company, Kia Motors Corporation
InventorsKyung Chul Lee, Jae Min Joh
Life of this patent
Estimating the term…
Abstract
A voice end-point detection device, a system and a method are provided. The voice end-point detection system includes a processor that is configured to determine an end-point detection time to detect an end-point of speaking of a user that varies vary for each user and for each domain. The voice end-point detection system is configured to perform voice recognition and a database (DB) is configured to store data for the voice recognition by the processor.
Claims19
A voice end-point detection system, comprising:
- a processor configured to set an end-point detection time for detecting an end-point of a voice signal spoken by a user among a plurality of users in a domain among a plurality of domains, the end-point detection time varying for each user and for each domain, and to perform voice recognition to recognize content of the voice signal;
- and
- a database (DB) configured to store data relating to the content of the voice signal recognized by the processor,
- wherein the processor is configured to identify the user and the domain upon the speaking of the user and determine a first end-point detection time and a second end-point detection time for each domain.
1 more element
A voice end-point detection device, comprising:
- an input signal managing unit configured to identify a user among a plurality of users and a domain for speech of the user among a plurality of domains and determine an end-point detection time for each user or for each domain using a per-user or per-domain end-point detection time DB;
- a voice detecting unit configured to detect voice signals spoken by the user by applying a plurality of end-point detection times for each user or for each domain;
- a voice recognition unit configured to perform voice recognition for each of the detected voice signals;
- a voice recognition result managing unit configured to search for each of the results of performing the voice recognition and output each of search;
- and
- an end-point detection managing unit configured to maintain or adjust the setting of the end-point detection time based on each of the search results,
- wherein the input signal managing unit is configured to identify the user and the domain upon the speaking of the user and determine a first end-point detection time and a second end-point detection tune for each domain, and
- wherein the first end-point detection time and the second end-point detection time are determined based on the user and are stored in the DB.
5 more elements
A voice end-point detection method, comprising:
- identifying, by a processor, a user among a plurality of users and a domain of the user among a plurality of domains and setting an end-point detection time determined for each user or for each domain using a per-user or per-domain end-point detection time DB;
- detecting, by the processor, voice signals spoken by the user by applying a plurality of end-point detection times determined for each user or for each domain;
- performing, by the processor, voice recognition for each of the detected voice signals;
- outputting, by the processor, each of search results by searching for each of the results of performing the voice recognition;
- and
- maintaining or adjusting, by the processor, the setting of the end-point detection time based on each of the search results,
- wherein the setting of an end-point detection time determined for each user or for each domain includes:
- identifying the user and the domain upon the speaking of the user;
- and
- determining a first end-point detection time and a second end-point detection time for each domain, and
- wherein the first end-point detection time and the second end-point detection time are determined based on the user and are stored in the DB.
8 more elements
Dependencies are inferred from the claim text.