{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T11:06:05Z","timestamp":1790766365136,"version":"4.1.0"},"reference-count":19,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Robot. AI"],"abstract":"<jats:p>Humanoid social robots deployed in multilingual societies\u2014particularly in indigenous communities\u2014face a critical linguistic gap: their speech interfaces are predominantly designed for high-resource languages, excluding hundreds of millions of indigenous language speakers from equitable human-robot interaction. This study presents the first neural text-to-speech (TTS) system for Southern Quechua\u2014the most extensive indigenous language family of the Americas, with more than ten million active speakers\u2014deployed on a humanoid social robot. XTTS-v2, a multilingual neural TTS model, was fine-tuned on 22,727 validated utterances (30.5 h, 77 speakers) of the Cusco-Puno dialect, located in southern Peru, through a tokenizer adaptation that leverages the phonological overlap between Quechua and Spanish. Fine-tuning over 15 epochs on an NVIDIA A100 40 GB GPU achieved a 35.0% reduction in training cross-entropy loss (4.084 \u2192 2.652), confirming effective adaptation to Quechua phonology and morphology with stable training dynamics across 35,000 gradient update steps. Inference latency was evaluated on two hardware environments: the training GPU (mean latency: 2,037 ms; Real-Time Factor (RTF) = 0.472) and the embedded SoC of the UBTech Walker Tienkung humanoid robot (Jetson AGX Orin 64 GB; mean latency: 5,913 ms; RTF = 0.961). Both configurations achieved a mean RTF below 1.0, demonstrating real-time Quechua speech synthesis directly on the robot\u2019s hardware without external computing infrastructure. Intelligibility evaluation using Whisper large-v3 with automatic language detection yielded a Word Error Rate of 147.4% and a Character Error Rate of 69.1%; these values reflect the absence of a native Quechua automatic speech recognition (ASR) model rather than the quality of TTS synthesis. These results establish the first quantitative benchmark for TTS in American indigenous languages on a humanoid robotic platform and provide a reproducible framework\u2014encompassing training configuration, tokenizer adaptation, and deployment pipeline\u2014for extending neural speech synthesis to other low-resource languages of the Americas.<\/jats:p>","DOI":"10.3389\/frobt.2026.1909076","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T10:42:18Z","timestamp":1790764938000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Enabling Quechua speech in humanoid social robots through fine-tuned neural TTS on the UBTech Walker Tienkung"],"prefix":"10.3389","volume":"13","author":[{"given":"Percy","family":"Castro-Mejia","sequence":"first","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William","family":"Atalaya-Urrutia","sequence":"additional","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alejandro","family":"Cruzata-Martinez","sequence":"additional","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erick","family":"Salazar-Montoya","sequence":"additional","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alberto","family":"Jimenez-Garcia","sequence":"additional","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan","family":"Arcila-Diaz","sequence":"additional","affiliation":[{"name":"Universidad Se\u00f1or de Sip\u00e1n","place":["Chiclayo, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,30]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"744526","DOI":"10.3389\/frobt.2021.744526","article-title":"10 years of Human-NAO interaction research: a scoping review","volume":"8","author":"Amirova","year":"2021","journal-title":"Front. Robotics AI"},{"key":"B2","first-page":"1","article-title":"You don\u2019t need to speak, you need to listen: robot interaction and human-like turn-taking","volume-title":"Proceedings of the 5th International Conference on Conversational User Interfaces","author":"Aylett","year":"2023"},{"key":"B3","doi-asserted-by":"publisher","first-page":"4978","DOI":"10.21437\/Interspeech.2024-2016","article-title":"XTTS: a massively multilingual zero-shot text-to-speech model","volume":"2024","author":"Casanova","year":"2024","journal-title":"Interspeech"},{"key":"B4","doi-asserted-by":"publisher","first-page":"466","DOI":"10.5753\/jbcs.2025.5468","article-title":"The evaluation of prosody in speech synthesis: a systematic review","volume":"31","author":"Galdino","year":"2025","journal-title":"J. Braz. Comput. Soc."},{"key":"B5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TENCON55691.2022.9977765","article-title":"Hindi spoken conversational system for healthcare humanoid","volume-title":"TENCON 2022 - 2022 IEEE Region 10 Conference (TENCON)","author":"Hegde","year":"2022"},{"key":"B6","first-page":"184","volume-title":"Quechua Speech Datasets in Common Voice: The Case of Puno Quechua","author":"Huaman","year":"2026"},{"key":"B7","doi-asserted-by":"crossref","first-page":"6282","DOI":"10.18653\/v1\/2020.acl-main.560","article-title":"The State and fate of linguistic diversity and inclusion in the NLP world","volume-title":"Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics","author":"Joshi","year":"2020"},{"key":"B8","first-page":"5530","article-title":"Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech","volume":"139","author":"Kim","year":"2021","journal-title":"Proc. Mach. Learn. Res."},{"key":"B9","doi-asserted-by":"publisher","first-page":"17022","DOI":"10.48550\/arXiv.2010.05646","article-title":"HiFi-GAN: generative adversarial networks for efficient and high fidelity speech synthesis","volume":"33","author":"Kong","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B10","unstructured":"Giving voice to the constitution: Low-Resource text-to-speech for Quechua and Spanish using a bilingual legal corpus\n          \n          \n            \n              Ortega\n              J. E.\n            \n            \n              Zevallos\n              R.\n            \n            \n              Carraro\n              F.\n            \n          \n          \n          2026"},{"key":"B11","doi-asserted-by":"publisher","first-page":"778","DOI":"10.1038\/s41597-022-01855-9","article-title":"A speech corpus of Quechua Collao for automatic dimensional emotion recognition","volume":"9","author":"Paccotacya-Yanque","year":"2022","journal-title":"Sci. Data"},{"key":"B12","doi-asserted-by":"publisher","first-page":"159","DOI":"10.15625\/1813-9663\/18136","article-title":"Adapt-TTS: High-quality zero-shot multi-speaker text-to-speech adaptive-based for Vietnamese","volume":"39","author":"Pham Ngoc","year":"2023","journal-title":"J. Comput. Sci. Cybern."},{"key":"B13","doi-asserted-by":"publisher","first-page":"101723","DOI":"10.1016\/j.csl.2024.101723","article-title":"Speech generation for Indigenous Language education","volume":"90","author":"Pine","year":"2025","journal-title":"Comput. Speech & Lang."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1032811","DOI":"10.3389\/frobt.2022.1032811","article-title":"Expansion in speech time can restore comprehension in a simultaneously speaking bilingual robot","volume":"9","author":"Pourfannan","year":"2023","journal-title":"Front. Robotics AI"},{"key":"B15","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","volume-title":"Proceedings of the 40th International Conference on Machine Learning","author":"Radford","year":"2023"},{"key":"B16","doi-asserted-by":"crossref","first-page":"4779","DOI":"10.1109\/ICASSP.2018.8461368","article-title":"Natural TTS synthesis by conditioning wavenet on MEL spectrogram predictions","volume-title":"2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","author":"Shen","year":"2018"},{"key":"B17","doi-asserted-by":"publisher","first-page":"101178","DOI":"10.1016\/j.csl.2020.101178","article-title":"Turn-taking in conversational systems and human-robot interaction: a review","volume":"67","author":"Skantze","year":"2021","journal-title":"Comput. Speech & Lang."},{"key":"B18","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1109\/ICEEICT62016.2024.10534321","article-title":"Disha: a bilingual humanoid virtual assistant","volume-title":"2024 6th International Conference on Electrical Engineering and Information &amp; Communication Technology (ICEEICT)","author":"Ullah","year":"2024"},{"key":"B19","doi-asserted-by":"publisher","first-page":"3518","DOI":"10.21437\/Interspeech.2022-770","article-title":"Data augmentation for low-resource Quechua ASR improvement","volume":"2022","author":"Zevallos","year":"2022","journal-title":"Interspeech"}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1909076\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T10:42:21Z","timestamp":1790764941000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1909076\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,30]]},"references-count":19,"alternative-id":["10.3389\/frobt.2026.1909076"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2026.1909076","relation":{},"ISSN":["2296-9144"],"issn-type":[{"value":"2296-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,30]]},"article-number":"1909076"}}