Speech technology for speech and hearing research

Speech technology can do much more than power voice assistants. In speech and hearing research and clinical practice, synthetic speech and automatic analysis make it possible to create controlled stimuli, measure speech at scale and build tools for people with communication difficulties.

How I can help

  • Stimulus creation with TTS: carefully controlled speech materials for listening experiments, where you can vary rate, pitch, prosody or voice while keeping everything else constant.
  • Speech-in-noise and perception tests: designing and generating test materials across voices and languages.
  • Automatic speech analysis: ASR, forced alignment and prosody measurement to analyze large speech corpora quickly and consistently.
  • ASR for atypical speech: evaluating and adapting recognition for speakers with speech disorders, hearing loss, or age-related changes in voice.
  • Experiment and evaluation design: listening tests and metrics that hold up to scientific review.

Who this is for

University labs, audiology and speech-language pathology researchers, hearing technology companies and clinical research teams. With my academic background, including my time at the Center for Spoken Language Understanding (CSLU) at OHSU, I am comfortable working within research projects, grants and publication timelines.