---
title: "Analyzing Autism Speech of Children in English Vowels Regions"
description: "The paper looks for acoustic traits that set apart the speech of children with autism spectrum disorder (ASD), an area with little… 24 references."
authors: ["Abhijit Mohanta", "Vinay Kumar Mittal"]
year: 2018
venue: "Proceedings of the 15th International Conference on Natural Language Processing"
source: "https://aclanthology.org/2018.icon-1.21/"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/mohanta-2018-analyzing-autism-speech-children-english/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/mohanta-2018-analyzing-autism-speech-children-english/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2018](https://md.research.elamur.ai/free/en/papers/2018/page/8/) / Analyzing Autism Speech of Children in English Vowels Regions by Analysis of Changes in Production Features

# Analyzing Autism Speech of Children in English Vowels Regions by Analysis of Changes in Production Features

**Authors:** Abhijit Mohanta and Vinay Kumar Mittal · **Year:** 2018 · **Venue:** [Proceedings of the 15th International Conference on Natural Language Processing](https://md.research.elamur.ai/free/en/venues/icon/) ([ICON 2018](https://md.research.elamur.ai/free/en/venues/icon/2018/)) · **Publisher:** [NLP Association of India](https://md.research.elamur.ai/free/en/publishers/nlp-association-of-india/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2018.icon-1.21/) · [PDF](https://aclanthology.org/2018.icon-1.21.pdf)

## Summary
The paper looks for acoustic traits that set apart the speech of children with autism spectrum disorder (ASD), an area with little prior work. The authors spent about a year building their own English recordings from 13 diagnosed children in Chennai. Each child repeatedly read a fixed list of picture-and-text prompts. The analysis covers only the five English vowel regions, because vowels last longer and are produced with quasi-periodic vocal-fold vibration. Features fall into three groups:
- **source:** F0, obtained with zero-frequency filtering (ZFF);
- **vocal-tract system:** dominant frequencies FD1 and FD2, from low-order linear prediction;
- **combined source-system:** signal energy, along with how strongly the source excites the tract (SoE) and how often the signal crosses zero (ZCR).

Mean values are compared between boys and girls for each vowel. Front and mid vowels follow similar trends for F0, E, SoE and ZCR in both sexes, while rear vowels do not. The authors suggest these patterns could help separate ASD speech from typical speech. However, no typically developing comparison group appears in the reported analysis, and only two girls were recorded.

## Key points
- **Dataset:** 13 children (11 male, 2 female), aged 3.5–16. Native languages were Tamil (12) and Punjabi (1), and English reading levels ranged from beginner to fluent. The corpus has 187 files totalling 9350 s, recorded on a Roland R-26 at 48 kHz in .wav format.
- **Elicitation:** 25 CVC/CVVC words and numbers, five per vowel, shown as pictures with text on a laptop. Recordings were made once or twice a week for a year in a quiet room.
- **Methods:** F0 was computed with ZFF at 10 kHz, using a cascade of two zero-frequency resonators and 10 ms trend removal. Glottal closure instants come from positive zero crossings, and SoE comes from the slope of the ZFF signal around those instants. FD1 and FD2 come from LP of order 5, which gives at most two spectral peaks. Energy used 30 ms frames with a 10 ms shift.
- **Findings by sex:** Girls show higher mean F0 for every vowel. Front vowels /e/ and /i/ give the highest F0 in both sexes, and boys have higher FD1 for all vowels.
- **Vowel /e/:** for girls it has the highest FD2 and the lowest FD1, while for boys it has the lowest FD2.

## Interesting ideas
- Analysis is limited to vowel stretches, where the signal is long and steady enough for reliable measurement. Restricting diagnostics to the most stable regions reduces noise from hard-to-measure parts.
- Setting the LP order to 5 limits the spectrum to at most two peaks, which turns a general model into two named, interpretable quantities (FD1, FD2). Deliberately limiting capacity can make a feature explainable by design.
- The same fixed stimulus set was repeated with the same speakers over a whole year, so any change reflects the speaker rather than the input.
- Features are grouped by where they originate (source, system, or their interaction), so an observed change can be traced to a component.
- The claim that features could distinguish ASD from typical speech rests on a single group with no reported control comparison, which shows why a baseline group is needed.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): compute parser and silver-data agreement first on constructions where witnesses are expected to agree, so that real drift is not hidden by inherently ambiguous cases.
- **I2** (world): compiled measures and values could be summarized by a small fixed number of named slots rather than open-ended vectors, so every derived quantity can be named in an explanation.
- **I3** (answer): keep a frozen probe set of school tests and fable questions that is re-run after every training round, so answer changes can be attributed to the model alone.
- **I4** (learn): label each error as caused by tokens, tree, frames or world, so repair rules target the stage that actually failed.
- **I5** (learn): every reported improvement should come with a negative control, such as shuffled or unrelated inputs, before it is treated as a real effect.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (2): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works

🧭 Topics (1): [English](https://md.research.elamur.ai/free/en/topics/english/) · 2,379 works

## Cite

Cite as: Abhijit Mohanta and Vinay Kumar Mittal. 2018. Analyzing Autism Speech of Children in English Vowels Regions by Analysis of Changes in Production Features. Proceedings of the 15th International Conference on Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{mohanta2018analyzing,
  title = {{Analyzing Autism Speech of Children in English Vowels Regions by Analysis of Changes in Production Features}},
  author = {Abhijit Mohanta and Vinay Kumar Mittal},
  booktitle = {Proceedings of the 15th International Conference on Natural Language Processing},
  year = {2018},
  url = {https://aclanthology.org/2018.icon-1.21/}
}
```

- [← previous An OpenNMT Model to Arabic Broken Plurals](https://md.research.elamur.ai/free/en/papers/issa-2018-opennmt-model-arabic-broken-plurals/)
- [next → Approche Hybride pour la translitération de l'Arabizi Algérien : une étude préliminaire (A hybrid approach for the transliteration of Algerian Arabizi: A primary study)](https://md.research.elamur.ai/free/en/papers/guellil-2018-approche-hybride-pour-la-transliteration/)
- [↑ back to papers of 2018](https://md.research.elamur.ai/free/en/papers/2018/page/8/)

## References

24 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Christiane AM Baltaxe. 1977. Pragmatic deficits in the language of autistic adolescents. Journal of Pediatric Psychology, 2(4):176–180. [find](https://scholar.google.com/scholar?q=Christiane+AM+Baltaxe%2E+1977%2E+Pragmatic+deficits+in+the+language+of+autistic+adolescents%2E+Journal+of+Pediatric+Psychology%2C+2%284%29%3A176%E2%80%93180%2E)
2. Matthew P Black, Daniel Bone, Marian E Williams, Phillip Gorrindo, Pat Levitt, and Shrikanth Narayanan. 2011. The usc care corpus: Child-psychologist interactions of children with autism spectrum disorders. In Twelfth Annual Conference of the International Speech Communication Association. [find](https://scholar.google.com/scholar?q=Matthew+P+Black%2C+Daniel+Bone%2C+Marian+E+Williams%2C+Phillip+Gorrindo%2C+Pat+Levitt%2C+and+Shrikanth+Narayanan%2E+2011%2E+The+usc+care+corpus%3A+Child%2Dpsychologist+interactions+of+children+with+autism+spectrum+diso)
3. Daniel Bone, Matthew P Black, Chi-Chun Lee, Marian E Williams, Pat Levitt, Sungbok Lee, and Shrikanth Narayanan. 2012. Spontaneous-speech acoustic-prosodic features of children with autism and the interacting psychologist. In Thirteenth Annual Conference of the International Speech Communication Association. [find](https://scholar.google.com/scholar?q=Daniel+Bone%2C+Matthew+P+Black%2C+Chi%2DChun+Lee%2C+Marian+E+Williams%2C+Pat+Levitt%2C+Sungbok+Lee%2C+and+Shrikanth+Narayanan%2E+2012%2E+Spontaneous%2Dspeech+acoustic%2Dprosodic+features+of+children+with+autism+and+the+int)
4. Theodora Chaspari, Matthew Goodwin, Oliver Wilder-Smith, Amanda Gulsrud, Charlotte A Mucchetti, Connie Kasari, and Shrikanth Narayanan. 2014. A non-homogeneous poisson process model of skin conductance responses integrated with observed regulatory behaviors for autism intervention. In Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on, pages 1611– 1615. IEEE. [find](https://scholar.google.com/scholar?q=Theodora+Chaspari%2C+Matthew+Goodwin%2C+Oliver+Wilder%2DSmith%2C+Amanda+Gulsrud%2C+Charlotte+A+Mucchetti%2C+Connie+Kasari%2C+and+Shrikanth+Narayanan%2E+2014%2E+A+non%2Dhomogeneous+poisson+process+model+of+skin+conductanc)
5. Joshua J Diehl, Duane Watson, Loisa Bennetto, Joyce McDonough, and Christine Gunlogson. 2009. An acoustic analysis of prosody in high-functioning autism. Applied Psycholinguistics, 30(3):385–404. [find](https://scholar.google.com/scholar?q=Joshua+J+Diehl%2C+Duane+Watson%2C+Loisa+Bennetto%2C+Joyce+McDonough%2C+and+Christine+Gunlogson%2E+2009%2E+An+acoustic+analysis+of+prosody+in+high%2Dfunctioning+autism%2E+Applied+Psycholinguistics%2C+30%283%29%3A385%E2%80%93404%2E)
6. Riccardo Fusaroli, Anna Lambrechts, Dan Bang, Dermot M Bowler, and Sebastian B Gaigg. 2017. Is voice a marker for autism spectrum disorder? a systematic review and meta-analysis. Autism Research, 10(3):384–407. [find](https://scholar.google.com/scholar?q=Riccardo+Fusaroli%2C+Anna+Lambrechts%2C+Dan+Bang%2C+Dermot+M+Bowler%2C+and+Sebastian+B+Gaigg%2E+2017%2E+Is+voice+a+marker+for+autism+spectrum+disorder%3F+a+systematic+review+and+meta%2Danalysis%2E+Autism+Research%2C+10%283)
7. Martha R Herbert et al. 2005. Autism: a brain disorder or a disorder that affects the brain. Clinical Neuropsychiatry, 2(6):354–379. [find](https://scholar.google.com/scholar?q=Martha+R+Herbert+et+al%2E+2005%2E+Autism%3A+a+brain+disorder+or+a+disorder+that+affects+the+brain%2E+Clinical+Neuropsychiatry%2C+2%286%29%3A354%E2%80%93379%2E)
8. Margaret M Kjelgaard and Helen Tager-Flusberg. 2001. An investigation of language impairment in autism: Implications for genetic subgroups. Language and cognitive processes, 16(2-3):287–308. [find](https://scholar.google.com/scholar?q=Margaret+M+Kjelgaard+and+Helen+Tager%2DFlusberg%2E+2001%2E+An+investigation+of+language+impairment+in+autism%3A+Implications+for+genetic+subgroups%2E+Language+and+cognitive+processes%2C+16%282%2D3%29%3A287%E2%80%93308%2E)
9. Catherine Lord, Michael Rutter, and Ann Le Couteur. 1994. Autism diagnostic interview-revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of autism and developmental disorders, 24(5):659–685. [find](https://scholar.google.com/scholar?q=Catherine+Lord%2C+Michael+Rutter%2C+and+Ann+Le+Couteur%2E+1994%2E+Autism+diagnostic+interview%2Drevised%3A+a+revised+version+of+a+diagnostic+interview+for+caregivers+of+individuals+with+possible+pervasive+develop)
10. Joanne McCann and Sue Peppé. 2003. Prosody in autism spectrum disorders: a critical review. International Journal of Language & Communication Disorders, 38(4):325–350. [find](https://scholar.google.com/scholar?q=Joanne+McCann+and+Sue+Peppe%CC%81%2E+2003%2E+Prosody+in+autism+spectrum+disorders%3A+a+critical+review%2E+International+Journal+of+Language+%26+Communication+Disorders%2C+38%284%29%3A325%E2%80%93350%2E)
11. Vinay Kumar Mittal and B Yegnanarayana. 2015a. Study of characteristics of aperiodicity in noh voices. The Journal of the Acoustical Society of America, 137(6):3411–3421. [find](https://scholar.google.com/scholar?q=Vinay+Kumar+Mittal+and+B+Yegnanarayana%2E+2015a%2E+Study+of+characteristics+of+aperiodicity+in+noh+voices%2E+The+Journal+of+the+Acoustical+Society+of+America%2C+137%286%29%3A3411%E2%80%933421%2E)
12. Vinay Kumar Mittal and Bayya Yegnanarayana. 2015b. Analysis of production characteristics of laughter. Computer Speech & Language, 30(1):99–115. [find](https://scholar.google.com/scholar?q=Vinay+Kumar+Mittal+and+Bayya+Yegnanarayana%2E+2015b%2E+Analysis+of+production+characteristics+of+laughter%2E+Computer+Speech+%26+Language%2C+30%281%29%3A99%E2%80%93115%2E)
13. Emily Mower, Matthew P Black, Elisa Flores, Marian Williams, and Shrikanth Narayanan. 2011a. Rachel: Design of an emotionally targeted interactive agent for children with autism. In Multimedia and Expo (ICME), 2011 IEEE International Conference on, pages 1–6. IEEE. [find](https://scholar.google.com/scholar?q=Emily+Mower%2C+Matthew+P+Black%2C+Elisa+Flores%2C+Marian+Williams%2C+and+Shrikanth+Narayanan%2E+2011a%2E+Rachel%3A+Design+of+an+emotionally+targeted+interactive+agent+for+children+with+autism%2E+In+Multimedia+and+Exp)
14. Emily Mower, Chi-Chun Lee, James Gibson, Theodora Chaspari, Marian E Williams, and Shrikanth Narayanan. 2011b. Analyzing the nature of eca interactions in children with autism. In Twelfth Annual Conference of the International Speech Communication Association. [find](https://scholar.google.com/scholar?q=Emily+Mower%2C+Chi%2DChun+Lee%2C+James+Gibson%2C+Theodora+Chaspari%2C+Marian+E+Williams%2C+and+Shrikanth+Narayanan%2E+2011b%2E+Analyzing+the+nature+of+eca+interactions+in+children+with+autism%2E+In+Twelfth+Annual+Confe)
15. K Sri Rama Murty and B Yegnanarayana. 2008. Epoch extraction from speech signals. IEEE Transactions on Audio, Speech, and Language Processing, 16(8):1602–1613. [find](https://scholar.google.com/scholar?q=K+Sri+Rama+Murty+and+B+Yegnanarayana%2E+2008%2E+Epoch+extraction+from+speech+signals%2E+IEEE+Transactions+on+Audio%2C+Speech%2C+and+Language+Processing%2C+16%288%29%3A1602%E2%80%931613%2E)
16. A Rihaczek. 1968. Signal energy distribution in time and frequency. IEEE Transactions on information Theory, 14(3):369–374. [find](https://scholar.google.com/scholar?q=A+Rihaczek%2E+1968%2E+Signal+energy+distribution+in+time+and+frequency%2E+IEEE+Transactions+on+information+Theory%2C+14%283%29%3A369%E2%80%93374%2E)
17. Michael Rutter. 1970. Autistic children: infancy to adulthood. In Seminars in psychiatry, volume 2, page 435. [find](https://scholar.google.com/scholar?q=Michael+Rutter%2E+1970%2E+Autistic+children%3A+infancy+to+adulthood%2E+In+Seminars+in+psychiatry%2C+volume+2%2C+page+435%2E)
18. Andrew B Short and Eric Schopler. 1988. Factors relating to age of onset in autism. Journal of autism and developmental disorders, 18(2):207–216. [find](https://scholar.google.com/scholar?q=Andrew+B+Short+and+Eric+Schopler%2E+1988%2E+Factors+relating+to+age+of+onset+in+autism%2E+Journal+of+autism+and+developmental+disorders%2C+18%282%29%3A207%E2%80%93216%2E)
19. Lawrence D Shriberg, Rhea Paul, Jane L McSweeny, Ami Klin, Donald J Cohen, and Fred R Volkmar. 2001. Speech and prosody characteristics of adolescents and adults with high-functioning autism and asperger syndrome. Journal of Speech, Language, and Hearing Research, 44(5):1097–1115. [find](https://scholar.google.com/scholar?q=Lawrence+D+Shriberg%2C+Rhea+Paul%2C+Jane+L+McSweeny%2C+Ami+Klin%2C+Donald+J+Cohen%2C+and+Fred+R+Volkmar%2E+2001%2E+Speech+and+prosody+characteristics+of+adolescents+and+adults+with+high%2Dfunctioning+autism+and+asper)
20. Helen Tager-Flusberg, Rhea Paul, Catherine Lord, F Volkmar, Rhea Paul, and Ami Klin. 2005. Language and communication in autism. Handbook of autism and pervasive developmental disorders, 1:335–364. [find](https://scholar.google.com/scholar?q=Helen+Tager%2DFlusberg%2C+Rhea+Paul%2C+Catherine+Lord%2C+F+Volkmar%2C+Rhea+Paul%2C+and+Ami+Klin%2E+2005%2E+Language+and+communication+in+autism%2E+Handbook+of+autism+and+pervasive+developmental+disorders%2C+1%3A335%E2%80%93364%2E)
21. John Makhoul. 1975. Linear prediction: A tutorial review. Proceedings of the IEEE, 63(4):561–580. [find](https://scholar.google.com/scholar?q=John+Makhoul%2E+1975%2E+Linear+prediction%3A+A+tutorial+review%2E+Proceedings+of+the+IEEE%2C+63%284%29%3A561%E2%80%93580%2E)
22. Fred R Volkmar, Kathy Koenig, and Matthew State. 1997. Childhood disintegrative disorder. Handbook of Autism and Pervasive Developmental Disorders, Volume 1, Third Edition, pages 70–87. [find](https://scholar.google.com/scholar?q=Fred+R+Volkmar%2C+Kathy+Koenig%2C+and+Matthew+State%2E+1997%2E+Childhood+disintegrative+disorder%2E+Handbook+of+Autism+and+Pervasive+Developmental+Disorders%2C+Volume+1%2C+Third+Edition%2C+pages+70%E2%80%9387%2E)
23. Erik Marchi, Björn W Schuller, Anton Batliner, Shimrit Fridenzon, Shahar Tal, and Ofer Golan. 2012. Emotion in the speech of children with autism spectrum conditions: prosody and everything else. In 149 WOCCI, pages 17–24. [find](https://scholar.google.com/scholar?q=Erik+Marchi%2C+Bjo%CC%88rn+W+Schuller%2C+Anton+Batliner%2C+Shimrit+Fridenzon%2C+Shahar+Tal%2C+and+Ofer+Golan%2E+2012%2E+Emotion+in+the+speech+of+children+with+autism+spectrum+conditions%3A+prosody+and+everything+else%2E+In+)
24. B Yegnanarayana and K Sri Rama Murty. 2009. Event-based instantaneous fundamental frequency estimation from speech signals. IEEE Transactions on Audio, Speech, and Language Processing, 17(4):614– 624. [find](https://scholar.google.com/scholar?q=B+Yegnanarayana+and+K+Sri+Rama+Murty%2E+2009%2E+Event%2Dbased+instantaneous+fundamental+frequency+estimation+from+speech+signals%2E+IEEE+Transactions+on+Audio%2C+Speech%2C+and+Language+Processing%2C+17%284%29%3A614%E2%80%93+624%2E)

*Updated 2 November 2026*