Hello!

I’m an Assistant Project Scientist in the Phonetics Lab (working with Georgia Zellou) in the UC Davis Department of Linguistics and a Specialist in the Vonk Lab (working with Jet Vonk) in the UCSF Department of Neurology and Memory and Aging Center. [bio]

My research examines how people communicate with and reason about artificial agents. I am particularly interested in how people produce, perceive, and learn speech patterns with other people vs. voice technology. I use these interactions as a window into cognition: how people form expectations about communicative partners, adapt their behavior during interaction, and respond to social cues from artificial agents.

I also study clinical translation, with a focus on cognitive aging. My work examines how AI systems can better support older adults in maintaining independence and how speech and language can serve as markers of cognitive change.

Broadly, my work sits at the intersection of psycholinguistics, phonetics, social cognition, and human-computer interaction. I combine experimental methods with computational analyses of speech to study both human and human-AI communication. For an overview of papers, see Research. You can also out some of my public science communication below: Public Outreach and Press!

Contact

mdcohn at ucdavis dot edu

michelle dot cohn at ucsf dot edu

she/her/hers

News

For all updates, see Recent posts

1. How do people talk to technology?

Our experiments have found that speakers adapt the acoustic properties of their speech (speech rate, pitch, intensity) when they are talking to a voice assistant, compared to another person. But many of the ways speakers adapt to local communicative contexts (e.g., a misunderstanding, mark focus, emotion) appear to be parallel in human-human and human-computer interaction.

  • Cohn, M., Barreda, S., Graf Estes, K., Yu, Z., & Zellou, G. (2024). Children and adults produce distinct technology- and human-directed speech. Scientific Reports. [OA article]
  • Cohn, M., Mengesha, Z., Lahav, M., & Heldreth, C. (2024). African American English speakersโ€™ pitch variation and rate adjustments for imagined technological and human addressees. Journal of Acoustical Society of America (JASA) Express Letters, 4(4). [OA Article]
  • Cohn, M., Bandodkar, G., Sangani, R., Predeck, K., & Zellou, G. (2024). Do People Mirror Emotion Differently with a Human or TTS Voice? Comparing Listener Ratings and Word Embeddings. Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. Honolulu, United States. [pdf][video][poster]
  • Vonessen, J., Aoki, N., Cohn, M., & Zellou, G. (2024). Comparing perception of L1 and L2 English by human listeners and machines: Effect of interlocutor adaptations. Journal of the Acoustical Society of America. [OA article]
  • Cohn, M., Ferenc Segedin, B., & Zellou, G. (2022). The acoustic-phonetic properties of Siri- and human-DS: Differences by error type and rate. Journal of Phonetics. [OA Article]
  • Beier, E., Cohn, M. (co-first authors), Trammel, T., Ferreira, F., & Zellou, G. (accepted). Marking Prosodic Prominence for Voice-AI and Human Addressees. Journal of Experimental Psychology: Learning, Memory, and Cognition. [PsyArXiv]
  • Cohn, M., & Zellou, G. (2021). Prosodic differences in human- and Alexa-directed speech, but similar error correction strategies. Frontiers in Communication. [OA Article]
  • Cohn, M., Liang, K., Sarian, M., Zellou, G., & Yu, Z. (2021). Speech rate adjustments in conversations with an Amazon Alexa socialbot. Frontiers in Communication [OA Article]
  • Cohn, M., Predeck, K., Sarian, M., & Zellou, G. (2021). Prosodic alignment toward emotionally expressive speech: Comparing human and Alexa model talkers. Speech Communication. [OA Article]
  • Perkins Booker, N., Cohn, M., & Zellou, G. (2024). Linguistic Patterning of Laughter in Human-Socialbot Interactions. Frontiers in Communication, 9, 1346738
  • Zellou, G., Cohn, M., & Kline, T. (2021). The Influence of Conversational Role on Phonetic Alignment toward Voice-AI and Human Interlocutors. Language, Cognition and Neuroscience [Article][pdf]
  • Cohn, M., & Zellou, G. (2019). Expressiveness influences human vocal alignment toward voice-AI. Interspeech [pdf]

2. How do people perceive artificial agents?

We’ve found that text-to-speech (TTS) generated voices carry important social information, including apparent age, gender, and emotion. People use this information to guide their interactions with technology, often applying the social ‘rules’ from human-human interaction. Yet, direct comparisons with human voices — or apparent human voices — shows that top-down expectations shape how TTS voices are perceived.

  • Cohn, M., Pushkarna, M., Olanubi, G., Moran, J., Padgett, D., Mengesha, Z., & Heldreth, C. (2024). Believing Anthropomorphism:  Examining the Role of Anthropomorphic Cues on Trust in Large Language Models. Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. Honolulu, United States. [pdf][video][poster]
  • Cohn, M., Bandodkar, G., Sangani, R., Predeck, K., & Zellou, G. (2024). Do People Mirror Emotion Differently with a Human or TTS Voice? Comparing Listener Ratings and Word Embeddings. Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. Honolulu, United States. [pdf]
  • Cohn, M., Pushkarna, M., Dรญaz, M., Cooper, J., Olanubi, F., Moran, J., Mengesha, Z., Heldreth, C. (2026). Age and generational differences in anthropomorphism and trust in large language models. Frontiers in Cognition, 5, 1824836. [paper]
  • Cohn, M. & Zellou, G. (2020). Perception of concatenative vs. Neural text-to-speech (TTS): Differences in intelligibility in noise and language attitudes. Interspeech [pdf] [Virtual Talk]
  • Aoki, N., Cohn, M., & Zellou, G. (2022). The clear speech intelligibility benefit for text-to-speech voices: Effects of speaking style and visual guise. Journal of Acoustical Society of America (JASA) Express Letters. [OA Article]
  • Cohn, M, Sarian, M., Predeck, K., & Zellou, G. (2020). Individual variation in language attitudes toward voice-AI: The role of listenersโ€™ autistic-like traits. Interspeech [pdf] [Virtual talk]
  • Zellou, G., Cohn, M., & Block, A. (2021). Partial compensation for coarticulatory vowel nasalization across concatenative and neural text-to-speech. Journal of the Acoustic Society of America [Article]
    Block, A., Cohn, M., & Zellou, G. (2021). Variation in Perceptual Sensitivity and Compensation for Coarticulation Across Adult and Child Naturally-produced and TTS Voices. Interspeech. [pdf]
  • Cohn, M., Raveh, E., Predeck, K., Gessinger, I., Mรถbius, B., & Zellou, G. (2020). Differences in Gradient Emotion Perception: Human vs. Alexa Voices. Interspeech [pdf] [Virtual talk]
  • Cohn, M., Chen, C., & Yu, Z. (2019). A Large-Scale User Study of an Alexa Prize Chatbot: Effect of TTS Dynamism on Perceived Quality of Social Dialog. SIGDial [pdf]
  • Gessinger, I., Cohn, M., Mรถbius, B., & Zellou, G (2022). Cross-Cultural Comparison of Gradient Emotion Perception: Human vs. Alexa TTS Voices. Interspeech [pdf].
  • Zhu, Q., Chau, A., Cohn, M., Liang, K-H, Wang, H-C, Zellou, G., & Yu, Z. (2022). Effects of Emotional Expressiveness on Voice Chatbot Interactions. 4th Conference on Conversational User Interfaces (CUI). [pdf]

3. How do people learn language from technology?

In human-human interaction, there are subconscious mechanisms that shape how we learn language. We learn the mapping between our accent and another speaker’s to understand a new phonological system. We also adopt the features of a language in our own speech. We’ve found that the type of talkerโ€” as an apparent human or a voice assistant โ€” shapes how people learn an accent and how they mirror another speaker’s pronunciation patterns.

  • Zellou, G., Cohn, M., & Pycha, A. (2023). The effect of listener beliefs on perceptual learning. Language. [pdf]
  • Ferenc Segedin, B. Cohn, M., & Zellou, G. (2019). Perceptual adaptation to device and human voices:  learning and generalization of a phonetic shift across real and voice-AI talkers. Interspeech [pdf]
  • Cohn, M., Keaton, K., Beskow, J., & Zellou, G. (2023). Vocal accommodation to technology: The role of physical form. Language Sciences 99, 101567. [OA Article]
  • Cohn, M., Jonell, P., Kim, T., Beskow, J., & Zellou, G. (2020). Embodiment and gender interact in alignment to TTS voices. Cognitive Science Society [OA Article] [Virtual talk]
  • Cohn, M., Ferenc Segedin, B., & Zellou, G. (2019). Imitating Siri: Socially-mediated vocal alignment to device and human voices. ICPhS [pdf]
  • Dodd, N., Cohn, M., & Zellou, G. (2023). Comparing alignment toward American, British, and Indian English text-to-speech (TTS) voices: Influence of social attitudes and talker guise. Frontiers in Computer Science, 5. [Article]
  • Zellou, G., Cohn, M., & Ferenc Segedin, B. (2021). Age- and gender-related differences in speech alignment toward humans and voice-AI. Frontiers in Communication [OA Article]
  • Zellou, G., & Cohn, M. (2020). Top-down effects of apparent humanness on vocal alignment toward human and device interlocutors. Cognitive Science Society [pdf]

4. Cognitive aging and language technology

As voice and AI technologies become part of daily life, speech and language offer a low-burden window into cognitive function. My work examines how technology can better support people as cognitive abilities change with aging and how features of speech and language can serve as markers of cognitive change.

At UCSF, I study acoustic and linguistic features associated with cognitive impairment, with the goal of identifying interpretable measures that could ultimately support earlier detection and clinical evaluation. I also examine how cognitive impairment affects interaction with voice technologies. For example, my work shows that older adults with dementia produce distinct speech patterns when talking to Alexa and are misunderstood more often by automatic speech recognition (ASR) systems than cognitively unimpaired adults.

More broadly, I collaborate with researchers in neurology, computer science, and engineering on aging-in-place technologies designed to support independence. This work includes conversational AI, accessible computer interfaces, remote cognitive monitoring, and evaluating how speech and language technologies can better accommodate changes in cognition and communication.

  • Challenges in Automatic Speech Recognition for Adults with Cognitive Impairment. (to appear). Cohn, M., Lanzi, A., Ishihara, Y., Chuah, C., Zellou, G., Weakley, A. Conference on Human Factors in Computing (CHI) 2026. Barcelona, Spain. [arxiv]
  • Koppolu, P., Gahoil, A., Patil, T., Yadav, P., Cohn, M., Chuah, C., Pan, S., Tomaszewski Farias, S., & Weakley, A. (2026)., Streamlining Remote Care: Computer Access Solutions for Older Adults with Cognitive Impairment. In: Tareq Ahram, Waldemar Karwowski and Jay Kalra (eds) Human Factors in Design, Engineering, and Computing. AHFE (2026) International Conference. AHFE Open Access, vol 199. AHFE International, USA.
  • Gohil, A., Bissig, D., Wang, H., Koppolu, P., Patil., T., Yadav, P., Pan, S., Cohn, M., Chuah, C., Tomaszewski Farias, S., & Weakley, A. (2026). Embedding self-administered dementia monitoring into a remote caregiving platform: A pilot usability and data-sharing preferences study. In: Tareq Ahram, Waldemar Karwowski and Jay Kalra (eds) Human Factors in Design, Engineering, and Computing. AHFE (2026) International Conference. AHFE Open Access, vol 199. AHFE International, USA.
  • Yadav, P., Agarwal., H., Koppolu, P., Gohil, A., Patil, T., Chuah, C., Pan, S., Aswar, S., Pimento, S., Cohn, M., & Weakley, A. (2026) Bounded Delegation of a Large Language Model in a Conversational Agent for Dementia Support. Proceedings of the IEEE AI x Medical Health, and Care (AIxMHC) Conference.
  • Yadav, P., Cohn, M., Koppolu, P., Agarwal, H., Gohil, A., Patil, T., Pimento, S., & Weakley, A. (2026). AI-Care: A Conversational Agentic System for Task Coordination in Alzheimerโ€™s Disease Care. IEEE/ACM CHASE Workshop: Generative AI for Smart Health and Biomedical Informatics.
  • Ikeuchi, K., Ohtsuki, Cohn, M., Weakley, A., Ishihara, Y., Hoang, J., Kishimoto, T., Tomaszewski Farias, S., Vonk, J., Chuah, C. (2026). Cross-linguistic acoustic biomarkers of cognitive impairment across interactive and single-speaker speech tasks. 48th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Toronto, Canada.

Back to top

Public Outreach

UC Davis Picnic Day: Speech Science Booth

Take our Children to Work (TOC) Day

  • 2023: We hosted families at the UC Davis Phonetics Lab for the 2023 Take Our Children to Work Day on April 27th. Children and adults participated in a real speech science experiment and learned more about the research we do at the PhonLab!
  • 2022: Virtual Event

2021 Picnic Day Virtual Booth
2020 Picnic Day Virtual Booth
2020 Picnic Day Virtual Booth

โ€œPhonologically motivated phonetic repair strategies in Siri- and human-directed speechโ€. in the 5 Minute Linguist competition at the 2019 Linguistic Society of America (LSA) annual meeting.  (Talk 3:23-7:31)

Back to top

Press

  • Across Acoustics [podcast]: “Why don’t speech recognition systems understand African American English?” (7/9/24)
  • New Scientist [article] “ChatGPT got an upgrade to make it seem more human” (5/14/24)
  • AIP Publishing [article] “Machine Listening: Making Speech Recognition Systems More Inclusive” (4/30/24)
  • Unfold Podcast [podcast] “Hey Siri, Why Do I Speak Differently to You?” (9/19/23)
  • StudyFinds [article]: AI baby talk? People change their voice while talking to Alexa and Siri (5/9/23)
  • EurekaAlert [article] Hey Siri, can you hear me? (5/9/23)
  • KCBS San Francisco (106.9FM AND 740AM)[article] [recording] Live Interview with Rebecca Corral “UC Davis conducts study on how wearing a mask affects our speech patternsโ€ (2/3/21)
  • CBS-13 Sacramento On-Air News Segment [article] [video] โ€œUC Davis Study Finds Face Masks Do Not Impact Ability To Communicateโ€ (2/2/21)
  • UC Davis Press Release [article] โ€œSpeaking and Listening Seem More Difficult in a Masked World, But People Are Adaptingโ€ (2/2/21)
  • In Focus [article]. “Speech Unmasked: UWM linguist studies how masks impact intelligibility” (3/2021)
  • CBS 58 Milwaukee On-Air News Segment [article/video]. “Researchers look at how masks impact communication (3/18/21)
  • Equinox [article] “Unmask Masked Speech” (3/6/21)
  • US News & World Report [article] โ€œAs Mask-Wearing Prevails, People Are Adapting to Understanding Speechโ€ (2/8/21)
  • Ladders [article] โ€œThis is how you can make masked conversations 100% more successfulโ€ (2/8/21)
  • WFMY Greensboro On-Air News Segment [article] [video] โ€œWhat did you say? How masks affect your communication & understandingโ€ (2/4/21)
Unfold Podcast: “Hey Siri, Why Do I Speak Differently to You?” (9/19/23)
CBS-13 Sacramento covered our recent face-masked speech paper on Feb. 2, 2021

Back to top