Spoken Communication Lab

The Spoken Communication Lab conducts research on early vocal development and speech and language acquisition in infants at risk of autism and related disorders.

The goal of the Spoken Communication Lab at Marcus Autism Center is to understand why children with autism spectrum disorders (ASD) have problems learning to talk, so that through a program of evidence-based research on early detection and targeted intervention we can make sure that no child with autism is left without a voice.

Why is research on speech important for understanding, detecting and treating autism? Deficits in social communication are one of the core features defining the disorder. Children with ASD are almost universally delayed in acquiring spoken language. Currently, about 25% of children with autism never learn to speak. Another 25% begin to produce sounds, but suffer an apparent regression between 12 and 18 months, with loss of first words and social withdrawal that may isolate them from their peers. The rest may achieve language by the time they reach adulthood, but they will continue to show severe difficulties with pragmatics in social situations and will often have noticeable differences in voice quality.

The link between autism and language ability

Delays and deviances in language acquisition are among the key diagnostic criteria for ASD. One of the main reasons parents of children with autism first seek help is because their child is not speaking properly by the time they reach their first birthday. At this age, most typically developing children would already be babbling and producing their first words. Language ability at 2 years old is the best prognostic indicator we have for long-term outcomes in autism in adulthood. Because speech is so important for so many aspects of a child’s learning and development, children with ASD who cannot speak by the time they are 2 years old are at risk for lifelong challenges.

However, the outlook is better today than it was was 20 years ago, when approximately 40% of children with autism reached adulthood without being able to speak. Nothing has changed about the condition, but advances in scientific research and clinical practice have led to earlier detection and better treatment, which we know is key to improving outcomes.

Seeking solutions through research

In order to build individualized intervention programs to help each child, we need to know more about what is going wrong with each child’s speech development. Because it is so difficult to track babies who go on to develop ASD prospectively, before they reach the age of diagnosis, right now we do not have the data to provide a complete picture of how speech develops in early infancy. We do not know why, or at what point, autism derails the acquisition of spoken language.

If we can develop objective quantitative methods for measuring the progress of each child from birth through the key stages of speech development that typical infants follow and catch the point at which they begin to falter, we can plan treatments in advance to help them over each hurdle before it occurs.

As part of a National Institutes of Health (NIH) Autism Center of Excellence funded by the National Institute of Mental Health and through academic partnerships between Children’s Healthcare of Atlanta, Emory University School of Medicine and Florida State University, our research program focuses on studying vocal development in ASD in the first three years of life. Thanks to the generous support of our donors, including The Marcus Foundation, the Joseph B. Whitehead Foundation and the Georgia Research Alliance, we are able to capitalize on state-of-the-art technologies to measure every aspect of a child’s behavior that may be relevant to social communication from the moment they are born and even when they are still in the womb.

Our hope is that through our science we may be able to detect signs of language delay or deviation from typical speech development at the earliest moment possible. If our research can identify the many potential ways in which the core deficits and comorbid conditions associated with ASD might impact the biological processes that should normally draw infants along the path to spoken language, we may be able to detect the derailment of spoken communication before it occurs. This will give us a window of opportunity to intervene as quickly and effectively as possible and help make sure every one of our children at risk of autism eventually learns to speak. At Marcus Autism Center, we believe that deficits in spoken communication are not an inevitable consequence of autism. Our aim is to change the lives of the children we serve by giving them back the power of speech or, better still, by making sure through the care we provide that they are never deprived of a voice at all.

Focus areas

  • Vocal biomarkers for early detection of autism and measurement of treatment response and outcome
  • Speech and language acquisition in ASD and related disorders
  • Speech science: speech production and perception from infancy through adulthood
  • Speech technology: automated analysis, synthesis and recognition of infant vocal behavior, adult speech and infant-caregiver interaction
  • Mathematical modeling: biomechanical, aerodynamic and acoustic simulations of sound production in the vocal tract, stochastic dynamical systems
  • Biostatistics: applied longitudinal data analysis, optimal nonuniform sampling strategies
  • Training and education: internship programs for high school, undergraduate, pre-doctoral and post-doctoral students

Associated studies

Prenatal Ultrasound Study (Emory IRB00054741)

Head and Brain Growth in Infancy Study (Emory IRB00064966)

Home Recordings of Infant Speech (Emory IRB00059383 & Emory IRB00097674; NIH P50 MH10009 & NIH R01 DC015108)

Recent Publications

  • Rodríguez-Saltos, C. A., Bhise, A., Karur, P., Khan, R. N., Lee, S., Ramsay, G. & Maney, D. L. (2023). Song preferences predict the quality of vocal learning in zebra finches. Scientific Reports 13(1):605.
  • Long, H. L., Ramsay, G., Griebel, U., Bene, E. R., Bowman, D. D., Burkhardt-Reed, M. M. & Oller, D. K. (2022). Perspectives on the origin of language: Infants vocalize most during independent vocal play but produce their most speech-like vocalizations during turn taking. PLoS One 17(12):e0279395.
  • Oller, D. K., Ramsay, G., Bene, E., Long, H. L. & Griebel, U. (2021). Speech-like sounds dominate the human infant vocal landscape. Philosophical Transactions of the Royal Society B 30(10):426-427.
  • Gipson, T., T., Ramsay, G., Ellison, E., Bene, E., Long, H. L. & Oller, D. K. (2021). Early vocal development in tuberous sclerosis complex. Pediatric Neurology 125:48-52.
  • Oller, D. K., Griebel, U., Bowman, D., Bene, E. R., Long, H. L., Yoo, H. & Ramsay, G., (2020). “Infant boys found to be more vocal than infant girls,” Current Biology 30:R417- R429.
  • Ramsay, G. J. (2019). Mechanical speech synthesis in early talking automata, Acoustics Today 15(2):11-19 (also selected for republication in special edition of Acoustics Today for the International Year of Sound in 2020).

Media coverage

“Signs of autism may be heard early in infancy.” INSAR 2019 presentation featured in Spectrum, May 2019.

“Internship gives teen with autism taste of success.” FOX 5 TV news interview featuring high-school internship program for adolescents with ASD, May 2019.

"Boys are vocal more often than girls in the first six months of life" Children's Healthcare of Atlanta press release.

Training opportunities

a. Postdoctoral Fellowships
Postdoctoral fellows interested in pursuing an academic or industrial career in speech science and engineering are welcome to collaborate on NIH-funded research on infant vocal development, speech production and perception, and early detection of risk in ASD.


  • Shweta Ghai, PhD

b. Marcus Fellowship in Speech Science and Engineering
As part of our NIH-funded research, we offer two-year predoctoral research fellowships designed to prepare graduate-level engineers and linguists for entry into a doctoral degree, leading either to an academic career in speech science and engineering or to a clinical career in speech-language pathology and communication disorders.


  • Kristin Muench
  • Yael Stern
  • Jhonelle Bailey
  • Rose-Milord Fleurissaint
  • Mitra Kumareswaran
  • Doroteja Dragovic

c. Marcus High School Internship in ASD Research
Through the generosity of our donors, we are able to offer flexible paid internships to high-school students each year, to enable them to collaborate on research projects with our faculty and staff. Priority is given to students diagnosed with ASD who have an interest in research.


  • Connor Samsky
  • Matthew Plunk
  • Isaiah Ortiz
  • Rushil Knagaram
  • Jayden Williams
  • Justin Delaine

d. ION Program in Neuroscience
For many years, we were a site for the Institute on Neuroscience (ION), an eight-week summer program encouraging high school students from under-represented minorities to take part in research, sponsored by the National Institute of Mental Health (NIMH) and hosted by Georgia State University, Emory University, and the Yerkes National Primate Research Center.


  • Rachel Schwartz
  • Catherine Benedict
  • Rama Balachandran
  • Sammi Gewirtz
  • Fahad Paryani

e. Undergraduate Research Fellows
Undergraduate students from Psychology, Linguistics, Engineering and Mathematics are welcome to conduct research projects in the lab on any aspect of speech science and engineering, either as part of their academic program or as summer projects.


  • Shivani Patel
  • Klaudia Bednarczyk
  • Crystal Kim
  • Sabrina Udwin
  • Sarah Lu
  • Alec Reinhardt

Gordon Ramsay, PhD
Assistant Professor,
Emory School of Medicine

Research coordinators
Kelcey Little
Mikafui Antonio

Former staff
Morgan Edwards

Contact information
Gordon Ramsay, PhD
Marcus Autism Center
1920 Briarcliff Road NE
Atlanta, GA 30329
Tel: 404-785-6085
Fax: 404-785-9083