Automatic Test Environment of Speech Recognition Software for Intelligent Display

Software test of smart display speech recognition software involves a combination of multiple external environmental noise, speed of speech, and intonation of instruction voice, which causes a large number of test case combinations and high execution costs.This paper proposed a test environment whic...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHONG Li, LIU Buqi, YANG Ying
Format: Article
Language:zho
Published: Editorial Office of Control and Information Technology 2020-01-01
Series:Kongzhi Yu Xinxi Jishu
Subjects:
Online Access:http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2020.04.007
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Software test of smart display speech recognition software involves a combination of multiple external environmental noise, speed of speech, and intonation of instruction voice, which causes a large number of test case combinations and high execution costs.This paper proposed a test environment which uses a script to automatically create and simultaneously play instruction voices with various types of elements combination, record and judge the recognition results. Core functions of the test environment are introduced and the system composition is described. Key technologies of the test environment are described, including multi-threaded speech synthesis and corresponding time performance test, and the execution sequence of a single test script is finally listed. The test environment can cover multiple combinations of speech elements of all voice instructions, and take into account the test requirements of environmental noise, good scalability and repeatability.
ISSN:2096-5427