IA, autisme et architecture de la voix : de l'exclusion conçue à la dignité par conception

Titre original en anglais : AI, autism, and the architecture of voice: from engineered exclusion to designed dignity

Sources de métadonnées utilisées pour cette référence

OpenAlex Crossref
En savoir plus sur les sources utilisées

Srinivasan, H. B. (2026). AI, autism, and the architecture of voice: from engineered exclusion to designed dignity. AI & SOCIETY. https://doi.org/10.1007/s00146-026-03044-3

Date de publication: 18/04/2026 Ajout dans AutiHub: 09/07/2026 Type: Article Langue de l’article: Anglais
Source OpenAlex Source éditeur (DOI) Texte intégral

Cette publication est intégrée dans AutiHub via :

Auteurs

Auteur·ices des publications
1
Auteur·ices de la publication identifié·es comme autistes
1 / 1 (100,0 %)
  1. Hari B. Srinivasan *

Résumé

Résumé Cet article conceptualise l'exclusion conçue — la mise à l'écart prévisible des utilisateurs handicapés résultant de choix relatifs à la provenance des données, aux objectifs des modèles et aux pratiques d'évaluation au sein des systèmes d'IA. Nous examinons l'exclusion conçue à travers les expériences vécues de personnes autistes minimalement verbales et non parlantes, dont les profils de communication remettent en question les paramètres par défaut centrés sur la parole intégrés aux pipelines d'IA contemporains. Ici, la « voix » désigne non seulement la parole, mais aussi l'architecture plus large par laquelle la communication incarnée et multimodale — couvrant le texte de CAA, les gestes, les mouvements et les vocalisations partielles — devient lisible au sein des systèmes d'IA. Le terme « non parlant » ne désigne donc pas l'absence de langage, mais un spectre hétérogène dans lequel la communication dépend souvent de l'état, variant selon la fatigue, l'anxiété, la charge sensorielle et les exigences de planification motrice — des formes de variation que les abstractions de conception effacent habituellement. En retraçant les mécanismes d'exclusion dans la reconnaissance vocale, les systèmes de synthèse vocale, les systèmes de langage clair et la conception d'interfaces, nous introduisons des métriques mesurables de dignité par conception pour l'évaluation technique (Tableau 1) et un cadre de gouvernance cartographiant la responsabilité tout au long du cycle de vie de l'IA (Tableau 2). Nous soutenons que l'accessibilité doit être traitée comme une dimension fondamentale de l'éthique de l'IA — au même titre que l'équité, la vie privée et la sécurité. Réingénierie de l'IA pour la dignité par conception exige des systèmes qui reconnaissent les formes incarnées, multimodales et fluctuantes de communication, élargissant ce qui est considéré comme un signal valide et une innovation responsable.

Abstract This paper conceptualizes engineered exclusion —the predictable sidelining of disabled users resulting from choices about data provenance, model objectives, and evaluation practices within AI systems. We examine engineered exclusion through the lived experiences of minimally and nonspeaking autistic people whose communicative profiles challenge the speech-centered defaults embedded in contemporary AI pipelines. Here, “voice” refers not only to speech but to the broader architecture through which embodied, multimodal communication—spanning AAC text, gesture, movement, and partial vocalizations—becomes legible within AI systems. “Nonspeaking” is therefore not the absence of language but a heterogeneous spectrum in which communication is often state dependent, varying with fatigue, anxiety, sensory load, and motor planning demands—forms of variation that design abstractions routinely erase. Tracing exclusionary mechanisms across speech recognition, text-to-speech, plain-language systems, and interface design, we introduce measurable designed-dignity metrics for technical evaluation (Table 1) and a governance framework mapping accountability across the AI lifecycle (Table 2). We argue that accessibility must be treated as a core dimension of AI ethics—on par with fairness, privacy, and safety. Re-engineering AI for designed dignity requires systems that recognize embodied, multimodal, and fluctuating forms of communication, expanding what counts as valid signal and responsible innovation.

Bibliographie citée par cette référence

Les références citées sont importées depuis des sources externes de métadonnées lorsqu’elles sont disponibles. La liste peut être partielle.

Vue d’ensemble de l’inclusion dans la bibliographie

Ces indicateurs décrivent la bibliographie citée importée pour cette publication. Les métriques de références citées utilisent le total des références citées comme dénominateur. Les métriques d’auteurices cité·es indiquent si elles utilisent toutes les occurrences d’auteurices cité·es ou seulement les occurrences rattachées à des auteurices déjà intégré·es à la base de données AutiHub. Ils utilisent les rattachements mis en cache entre les auteurices cité·es et les auteurices intégré·es à la base de données AutiHub. Dernier calcul : 16/08/2026 11:31.

Références citées
63
Nombre total de références citées intégrées pour cette publication.
Références citées avec un·e auteur·ice identifié·e comme autiste
4 / 63 (6,3 %)
Occurrences d’auteur·ices cité·es identifié·es comme autistes
8 / 157 (5,1 %)
Parmi les occurrences rattachées à des auteurices intégré·es à la base de données AutiHub : 8 / 23 (34,8 %). Auteurices cité·es distinct·es identifié·es comme autistes : 6 / 151 (4,0 %).
Occurrences citées rattachées à la base AutiHub
23 / 157 (14,6 %)
Auteurices cité·es distinct·es rattaché·es : 21 / 151 (13,9 %)
Occurrences rattachées, non identifiées comme autistes
15 / 23 (65,2 %)
Parmi les seules occurrences rattachées. Sur l’ensemble des occurrences d’auteurices cité·es : 15 / 157 (9,6 %). Auteurices cité·es distinct·es rattaché·es, non identifié·es comme autistes : 15 / 21 (71,4 %).
2 entrées de bibliographie citée n’ont pas pu être entièrement enrichies à partir des métadonnées DOI.
Crossref: introuvable (1) · DataCite: introuvable (1)
  1. Mary Wickenden (2011). Whose Voice is That?: Issues of Identity, Voice and Representation Arising in an Ethnographic Study of the Lives of Disabled Teenagers who use Augmentative and Alternative Communication (AAC) . Disability Studies Quarterly, 31(4). The Ohio State University Libraries.
    Type: Article DOI: 10.18061/dsq.v31i4.1724 OpenAlex: https://openalex.org/W1531676456
    Crossref OpenAlex
  2. Meryl Alper (2017). Giving Voice . The MIT Press.
    Type: Livre DOI: 10.7551/mitpress/10771.001.0001 OpenAlex: https://openalex.org/W4254014975
    Crossref OpenAlex
  3. Helen Tager‐Flusberg , Connie Kasari (2013). Minimally Verbal School‐Aged Children with Autism Spectrum Disorder: The Neglected End of the Spectrum . Autism Research, 6(6), 468-478. Wiley.
    Type: Article DOI: 10.1002/aur.1329 OpenAlex: https://openalex.org/W1909988460
    Crossref OpenAlex
  4. Fernando Alva-Manchego , Carolina Scarton , Lucia Specia (2020). Data-Driven Sentence Simplification: Survey and Benchmark . Computational Linguistics, 46(1), 135-187. MIT Press.
    Type: Article DOI: 10.1162/coli_a_00370 OpenAlex: https://openalex.org/W2997833324
    Crossref OpenAlex
  6. American Psychiatric Association (2022). Diagnostic and Statistical Manual of Mental Disorders . American Psychiatric Association Publishing.
    Type: Livre DOI: 10.1176/appi.books.9780890425787 OpenAlex: https://openalex.org/W4281687410
    Crossref OpenAlex
  7. Elizabeth B. Torres , Maria Brincker , Robert W. Isenhower , Polina Yanovich , Kimberly A. Stigler , John I. Nurnberger et al. (2013). Autism: the micro-movement perspective . Frontiers in Integrative Neuroscience, 7, 32-32. Frontiers Media SA.
    Type: Article DOI: 10.3389/fnint.2013.00032 OpenAlex: https://openalex.org/W2014585840
    Crossref OpenAlex
  8. (s. d.). Apple, n.d. Personal voice: create a voice that sounds like you. Apple Accessibility Support. https://support.apple.com . Accessed 2 Jan 2026
    Type: Autre
    Crossref
  9. Laura Schreibman , Géraldine Dawson , Aubyn C. Stahmer , Rebecca Landa , Sally J. Rogers , Gail G. McGee et al. (2015). Naturalistic Developmental Behavioral Interventions: Empirically Validated Treatments for Autism Spectrum Disorder . Journal of Autism and Developmental Disorders, 45(8), 2411-2428. Springer Science and Business Media LLC.
    Type: Article DOI: 10.1007/s10803-015-2407-8 OpenAlex: https://openalex.org/W2022662741
    Crossref OpenAlex
  10. Emily M. Bender , Timnit Gebru , Angelina McMillan-Major , Shmargaret Shmitchell (2021). On the Dangers of Stochastic Parrots . Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610-623. ACM.
    Type: Article DOI: 10.1145/3442188.3445922 OpenAlex: https://openalex.org/W3133702157
    Crossref OpenAlex
  11. Ilan Dinstein , David J. Heeger , Lauren Lorenzi , Nancy J. Minshew , Rafael Malach , Marlene Behrmann (2012). Unreliable Evoked Responses in Autism . Neuron, 75(6), 981-991. Elsevier BV.
    Type: Article DOI: 10.1016/j.neuron.2012.07.026 OpenAlex: https://openalex.org/W2133645220
    Crossref OpenAlex
  12. R Benjamin (2019). Race after technology: abolitionist tools for the New Jim Code . Race after technology: abolitionist tools for the New Jim Code.
    Type: Autre
    Crossref
  13. M A Dziuk , J C Gidley Larson , A Apostu , E M Mahone , M B Denckla , Stewart H. Mostofsky (2007). Dyspraxia in autism: association with motor, social, and communicative deficits . Developmental Medicine & Child Neurology, 49(10), 734-739. Wiley.
    Type: Article DOI: 10.1111/j.1469-8749.2007.00734.x OpenAlex: https://openalex.org/W2141607806
    Crossref OpenAlex
  14. D Beukelman (2020). Augmentative and alternative communication: supporting children and adults with complex communication needs . Augmentative and alternative communication: supporting children and adults with complex communication needs.
    Type: Autre
    Crossref
  15. Janice Light , David McNaughton (2012). Supporting the Communication, Language, and Literacy Development of Children with Complex Communication Needs: State of the Science and Future Research Priorities . Assistive Technology, 24(1), 34-44. Informa UK Limited.
    Type: Article DOI: 10.1080/10400435.2011.648717 OpenAlex: https://openalex.org/W2160965701
    Crossref OpenAlex
  16. Chitralekha Bhat , Helmer Strik (2025). Speech Technology for Automatic Recognition and Assessment of Dysarthric Speech: An Overview . Journal of Speech, Language, and Hearing Research, 68(2), 547-577. American Speech Language Hearing Association.
    Type: Article DOI: 10.1044/2024_jslhr-23-00740 OpenAlex: https://openalex.org/W4406385914
    Crossref OpenAlex
  17. Connie Kasari , Nancy Brady , Catherine Lord , Helen Tager‐Flusberg (2013). Assessing the Minimally Verbal School‐Aged Child With Autism Spectrum Disorder . Autism Research, 6(6), 479-493. Wiley.
    Type: Article DOI: 10.1002/aur.1334 OpenAlex: https://openalex.org/W2165555242
    Crossref OpenAlex
  18. Nancy C. Brady , Susan Bruce , Amy Goldman , Karen Erickson , Beth Mineo , Bill T. Ogletree et al. (2016). Communication Services and Supports for Individuals With Severe Disabilities: Guidance for Assessment and Intervention . American Journal on Intellectual and Developmental Disabilities, 121(2), 121-138. Informa UK Limited.
    Type: Article DOI: 10.1352/1944-7558-121.2.121 OpenAlex: https://openalex.org/W2277649489
    Crossref OpenAlex
  19. Janice Light , David McNaughton (2013). Putting People First: Re-Thinking the Role of Technology in Augmentative and Alternative Communication Intervention . Augmentative and Alternative Communication, 29(4), 299-309. Informa UK Limited.
    Type: Article DOI: 10.3109/07434618.2013.848935 OpenAlex: https://openalex.org/W2167196699
    Crossref OpenAlex
  20. Center for Democracy & Technology (2022). Ableism and disability discrimination in new surveillance technologies . Ableism and disability discrimination in new surveillance technologies.
    Type: Autre
    Crossref
  21. Gabriel J. Cler , Katharine R. Kolin , Jacob P. Noordzij , Jennifer M. Vojtech , Susan K. Fager , Cara E. Stepp (2019). Optimized and Predictive Phonemic Interfaces for Augmentative and Alternative Communication . Journal of Speech, Language, and Hearing Research, 62(7), 2065-2081. American Speech Language Hearing Association.
    Type: Article DOI: 10.1044/2019_jslhr-s-msc18-18-0187 OpenAlex: https://openalex.org/W2957373230
    Crossref OpenAlex
  22. Dirk Hovy , Shannon L. Spruit (2016). The Social Impact of Natural Language Processing . Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 591-598. Association for Computational Linguistics.
    Type: Article DOI: 10.18653/v1/p16-2096 OpenAlex: https://openalex.org/W2511234952
    Crossref OpenAlex
  23. (s. d.). CommunicationFIRST, n.d. About nonspeaking people. Available at: https://communicationfirst.org/ . Accessed 2 Jan 2026
    Type: Autre
    Crossref
  24. Brett Heasman , Alex Gillespie (2018). Perspective-taking is two-sided: Misunderstandings between people with Asperger’s syndrome and their family members . Autism, 22(6), 740-750. SAGE Publications.
    Type: Article DOI: 10.1177/1362361317708287 OpenAlex: https://openalex.org/W2727105378
    Crossref OpenAlex
  25. Coqui AI (2024) XTTS-v2: cross-lingual voice cloning with minimal reference audio. Technical documentation
    Type: Autre
    Crossref
  26. Sasha Costanza-Chock (2020). Design Justice . The MIT Press.
    Type: Livre DOI: 10.7551/mitpress/12255.001.0001 OpenAlex: https://openalex.org/W4252148582
    Crossref OpenAlex
  27. Jason Taylor , Korin Richmond (2019). Analysis of Pronunciation Learning in End-to-End Speech Synthesis . Interspeech 2019, 2070-2074. ISCA.
    Type: Article DOI: 10.21437/interspeech.2019-2830 OpenAlex: https://openalex.org/W2972610613
    Crossref OpenAlex
  28. Deviyani A (2022) Text normalization and its role in speech synthesis. In: Proceedings of the S4SG 2022 Workshop
    Type: Autre
    Crossref
  29. Allison Koenecke , Andrew Nam , Emily Lake , Joe Nudell , Minnie Quartey , Zion Mengesha et al. (2020). Racial disparities in automated speech recognition . Proceedings of the National Academy of Sciences, 117(14), 7684-7689. National Academy of Sciences.
    Type: Article DOI: 10.1073/pnas.1915768117 OpenAlex: https://openalex.org/W3012624518
    Crossref OpenAlex
  30. ElevenLabs (2025) Impact program: accessible voice technology. ElevenLabs Documentation
    Type: Autre
    Crossref
  31. Janna van Grunsven , Sabine Roeser (2022). AAC Technology, Autism, and the Empathic Turn . Social Epistemology, 36(1), 95-110. Informa UK Limited.
    Type: Article DOI: 10.1080/02691728.2021.1897189 OpenAlex: https://openalex.org/W3154144522
    Crossref OpenAlex
  32. Elizabeth Ellcessor (2016). Restricted Access . New York University Press.
    Type: Livre DOI: 10.18574/nyu/9781479867431.001.0001 OpenAlex: https://openalex.org/W4393292492
    Crossref OpenAlex
  33. Aimi Hamraie (2017). Building Access . University of Minnesota Press.
    Type: Livre DOI: 10.5749/minnesota/9781517901639.001.0001 OpenAlex: https://openalex.org/W4247616855
    Crossref OpenAlex
  34. European Parliament , Council of the European Union (2024). European Parliament and Council of the European Union (2024) Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 on artificial intelligence and amending certain Union legislative acts (Artificial Intelligence Act). Off J Eur Union L168:1–158 Off J Eur Union, L168, 1.
    Type: Autre
    Crossref
  35. FishAudio (2025) S1-mini multilingual TTS model documentation
    Type: Autre
    Crossref
  36. Nils Freyer , Hendrik Kempt , Lars Klöser (2024). Easy-read and large language models: on the ethical dimensions of LLM-based text simplification . Ethics and Information Technology, 26(3). Springer Science and Business Media LLC.
    Type: Article DOI: 10.1007/s10676-024-09792-4 OpenAlex: https://openalex.org/W4401315191
    Crossref OpenAlex
  37. Grand View Research (2024). Voice and speech recognition market size, share & trends report, 2024–2030 . Voice and speech recognition market size, share & trends report, 2024–2030.
    Type: Autre
    Crossref
  38. Zhaopeng Qian , Kejing Xiao (2023). A Survey of Automatic Speech Recognition for Dysarthric Speech . Electronics, 12(20), 4278. MDPI AG.
    Type: Article DOI: 10.3390/electronics12204278 OpenAlex: https://openalex.org/W4387672817
    Crossref OpenAlex
  39. Kerri Prinos , Neal Patwari , Cathleen A. Power (2024). Speaking of accent: A content analysis of accent misconceptions in ASR research . The 2024 ACM Conference on Fairness, Accountability, and Transparency, 1245-1254. ACM.
    Type: Article DOI: 10.1145/3630106.3658969 OpenAlex: https://openalex.org/W4399362524
    Crossref OpenAlex
  40. S Hendren (2020). What can a body do? How we meet the built world . What can a body do? How we meet the built world.
    Type: Autre
    Crossref
  41. T Iacono (2018). Iacono T, Trembath D, Erickson S (2018) The role of augmentative and alternative communication for children with autism: current status and future trends. Neuropsychiatr Dis Treat 14:3049–3061 Neuropsychiatr Dis Treat, 14, 3049.
    Type: Autre
    Crossref
  42. Alicia Martin , Robert L. MacDonald , Pan-Pan Jiang , Marilyn Ladewig , Julie Cattiau , Rus Heywood et al. (2025). Project Euphonia: advancing inclusive speech recognition through expanded data collection and evaluation . Frontiers in Language Sciences, 4. Frontiers Media SA.
    Type: Article DOI: 10.3389/flang.2025.1569448 OpenAlex: https://openalex.org/W4411500951
    Crossref OpenAlex
  44. Vishnu Raja , Adithya V Ganesan , Anand Syamkumar , Ritwik Banerjee , H. Schwartz (2025). Idiosyncratic Versus Normative Modeling of Atypical Speech Recognition: Dysarthric Case Studies . Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 33514-33525. Association for Computational Linguistics.
    Type: Article DOI: 10.18653/v1/2025.emnlp-main.1701 OpenAlex: https://openalex.org/W4416035306
    Crossref OpenAlex
  45. K Kuligowska (2018). Kuligowska K, Kisielewicz P, Włodarz A (2018) Speech synthesis systems: disadvantages and limitations. Int J Eng Technol (IJET) 7(2. 28):234–239. https://doi.org/10 . 14419/ijet.v7i2. 28. 12356 Int J Eng Technol (IJET), 7(2.28), 234.
    Type: Autre DOI: 10.14419/ijet.v7i2.28.12356 Crossref: introuvable DataCite: introuvable
    Crossref
  46. Emily D. Quinn , Kim Kurin , Alexandria Cook , Mari Therrien , Mollie Romano (2025). Experiences and Contextual Factors Impacting AAC Use at Home: A Qualitative Interview Study . Topics in Early Childhood Special Education. SAGE Publications.
    Type: Article DOI: 10.1177/02711214251389895 OpenAlex: https://openalex.org/W4416273041
    Crossref OpenAlex
  49. Lopez Lloreda C (2020) Speech recognition tech is yet another example of bias. Scientific American. Available at: https://www.scientificamerican.com/article/speech-recognition-tech-is-yet-another-example-of-bias/
    Type: Autre
    Crossref
  50. F Masina (2020). Masina F, Orso V, Pluchino P, De Angeli A (2020) Investigating the accessibility of voice assistants with impaired users: a mixed-methods study. Int J Hum Comput Interact 36(12):1153–1169 Int J Hum Comput Interact, 36(12), 1153.
    Type: Autre
    Crossref
  51. LL Piepzna-Samarasinha (2018). Care work: Dreaming disability justice . Care work: Dreaming disability justice.
    Type: Autre
    Crossref
  52. (s. d.). Plain Language Action & Information Network, n.d., What is plain language? US Government. Available at: https://www.plainlanguage.gov/about/definitions/ . Accessed 20 Dec 2025
    Type: Autre
    Crossref
  53. BM Prizant (2006). The SCERTS model: a comprehensive educational approach for children with autism spectrum disorders . The SCERTS model: a comprehensive educational approach for children with autism spectrum disorders.
    Type: Autre
    Crossref
  54. Rae JW, Potapenko A, Jayakumar SM et al (2021) Scaling language models: Methods, analysis & insights from training Gopher. arXiv preprint arXiv:2112. 11446
    Type: Autre
    Crossref
  55. Reichel UD, Pfitzinger H (2006) Text preprocessing for speech synthesis, In: Text, speech and dialogue: 9th international conference, TSD 2006, Springer, Berlin, pp. 185–192
    Type: Autre
    Crossref
  56. Sharma N, Murray K and Xiao Z (2024) Faux polyglot: a study on information disparity in multilingual large language models. arXiv preprint arXiv:2407. 05502
    Type: Autre
    Crossref
  57. Shor J, Jansen A, Heck L, Simko J (2019) Towards learning speaker-invariant representations for dysarthric speech recognition. In: Proceedings of interspeech 2019, pp. 4564–4568
    Type: Autre
    Crossref
  58. Srinivasan H (2023) Flexible work critics are using the same arguments that were used to oppose disabled ramps and closed captioning—equity of access should never be optional, Fortune. Available at: https://fortune.com/2023/07/27/flexible-work-critics-using-same-arguments-were-used-oppose-disabled-ramps-closed-captioning-equity-access-never-optional-remote-work-careers-hari-srinivasan/
    Type: Autre
    Crossref
  59. K Trnka (2008). Trnka K, McCaw J, Yarrington D, McCoy KF, Pennington C (2008) Word prediction and communication rate in AAC. Technol Disabil 20(3):191–199 Technol Disabil, 20(3), 191.
    Type: Autre
    Crossref
  60. US General Services Administration (2023) Section508 of the Rehabilitation Act of 1973 (29 U. S. C. § 794d): Information and communication technology accessibility standards, US GSA, Washington, DC. Available at: https://www.section508.gov/manage/laws-and-policies/
    Type: Autre
    Crossref
  61. US Office of Management , Budget (2011). Final guidance on implementing the Plain Writing Act of 2010 . Final guidance on implementing the Plain Writing Act of 2010.
    Type: Autre
    Crossref
  62. M Vargas (2019). Vargas M (2019) Design and evaluation of a context-adaptive AAC system. ACM SIGACCESS Access Comput 124:Article 2 ACM SIGACCESS Access Comput, 124.
    Type: Autre
    Crossref
  63. M Whittaker (2021). Disability, bias, and AI . Disability, bias, and AI.
    Type: Autre
    Crossref