Synthesizing Speech Test Cases with Text-to-Speech? An Empirical Study on the False Alarms in Automated Speech Recognition Testing.
Julia Kaiwen LauKelvin Kai Wen KongJulian Hao YongPer Hoong TanZhou YangZi Qian YongJoshua Chern Wey LowChun Yong ChongMei Kuan LimDavid LoPublished in: ISSTA (2023)
Keyphrases
- test cases
- false alarms
- speech synthesis
- text to speech
- speech recognition
- software testing
- prosodic features
- detection rate
- testing process
- test data
- test suite
- test case generation
- vocal tract
- model based testing
- regression testing
- number of test cases
- test data generation
- test suite reduction
- test generation
- test case selection
- speech signal
- speech processing
- text to speech synthesis
- automatic speech recognition
- false positives
- language model
- pattern recognition
- hidden markov models
- set of test cases
- speech recognizer
- test set
- speaker identification
- speech recognition systems
- object oriented
- english text
- speech recognition technology
- noisy environments
- speech recognition errors
- acoustic models
- word processing