Independent expertise for teams building voice AI
I help companies build, improve and evaluate speech technology: text-to-speech, speech recognition and voice agents. I bring many years of research and industry experience in speech and language technology, with a focus on quality you can measure.
- Former Senior Research Scientist, ReadSpeaker
- Vice-President, ISCA (International Speech Communication Association)
- Adjunct Lecturer, Pacific University and Willamette University
- Technical Program Chair, Interspeech 2025
- Former Assistant Professor, CSLU, OHSU
- Former Chief Engineer, Biospeech
How I can help
Speech synthesis (TTS)
Text normalization, pronunciation and G2P, homograph disambiguation, prosody, multilingual voices and ethically sourced speech data.
Speech recognition (ASR)
Test set design, error analysis, domain and accent robustness, and choosing or adapting the right model for your users.
Voice agents & evaluation
End-to-end quality audits for voice assistants and agents: recognition, pronunciation, latency, turn-taking and what users actually experience.
Focus areas
Beyond commercial products, I am especially interested in projects where speech technology can make a difference for people who are often left out:
- Under-resourced and Indigenous languages: TTS and ASR built in partnership with language communities.
- Speech and hearing research: synthetic stimuli, automatic speech analysis and ASR for atypical speech.
Ways to work together
- Voice quality audit: a focused review of your TTS, ASR or voice agent with a prioritized list of issues and fixes.
- Evaluation framework: test sets, metrics and listening tests designed around your product and your users.
- Data collection & annotation: speech and text datasets built with informed consent and fair compensation.
- Advisory: ongoing expert input for your team on research direction, hiring and technical decisions.
Latest from the blog
- Interspeech 2026 in Sydney, AustraliaBack in Sydney 28 years after ICSLP 1998: highlights from Interspeech 2026, including its focus on Indigenous and Pacific voices and on speech technology for speech and hearing sciences.
- Homograph DisambiguationWe are currently at the stage where AI text-to-speech (TTS) can generate voices that sound as natural as human speech. There are many use cases out there and many AI companies offering these voices in Read more
- Interspeech 2025 RotterdamWhen I was asked to be Technical Program Chair for Interspeech 2025 in Rotterdam, I had no idea of the amount of work involved in this task. But after months of hard work, together with Read more