150
AUTHENTIC TOURISM TASKS + AI SPEAKING
STUDIO: A SCALABLE MODEL FOR FLUENCY,
INTELLIGIBILITY, AND TRANSFER
TAREAS DE TURISMO AUTÉNTICO + ESTUDIO DE
HABLA CON IA: UN MODELO ESCALABLE PARA LA
FLUIDEZ, LA INTELIGIBILIDAD Y LA TRANSFERENCIA
Recibido: 28/05/2025 - Aceptado: 13/01/2026
Claudia Verónica Sánchez Díaz
Docente de la Universidad de Ciencias y Artes de Chiapas
México
Especialidad en Enseñanza del Inglés
Universidad Tangamanga
claudia.sanchezd@unicach.mx
https://orcid.org/0009-0000-2171-0704
Como citar este artículo:
Sánchez, C. (Enero – junio 2026). Authentic tourism tasks + ai speaking studio: a scalable
model for uency, intelligibility, and transfer. Sathiri, 21 (1), 150 – 154.
https://doi.org/10.32645/13906925.1449
151
Authentic Tourism Tasks + AI Speaking
Studio: A Scalable Model for Fluency,
Intelligibility, and Transfer
Sánchez, C. (Enero – junio 2026). Authentic tourism tasks + ai speaking studio: a scalable model for uency, intelligibility, and transfer. Sathiri, 21
(1), 150 – 154. https://doi.org/10.32645/13906925.1449
Resumen
Este artículo presentó el diseño e implementación inicial de un ciclo de seis semanas de Aprendizaje
Basado en Tareas (TBLT) integrado con un ‘Speaking Studio’ asistido por IA para estudiantes de tercer
semestre de Turismo Sustentable (n=20; 10 horas/semana) en Palenque, México, en colaboración
con el Hotel Misión Palenque. El objetivo fue mejorar la uidez, inteligibilidad y transferencia
del inglés a interacciones reales de servicio. La intervención combinó tareas auténticas (un pitch
de conserjería de 90 segundos y un microartefacto bilingüe) con una rutina semanal de práctica
deliberada (imitar–grabar–reexionar) apoyada por ChatGPT para modelado de géneros, bancos de
frases, detección de problemas de claridad y generación de micro-quizzes de repaso. Se planicó una
evaluación formativa continua (salidas rápidas, bitácoras de errores) y una sumativa con rubrica de
cuatro bandas para inteligibilidad, uidez, adecuación e interacción, complementada por métricas
derivadas de transcripciones (palabras por minuto, longitud media de enunciado y variedad léxica),
valoración del socio comunitario y una escala breve de ansiedad. Los resultados esperados incluyeron
aumentos en ritmo con precisión estable, mejoras en acentuación y movimientos discursivos de
servicio, mayor conanza para atender visitantes y adopción de al menos un artefacto por parte del
hotel. El modelo demostró ser escalable, de bajo costo y reproducible, aliviando el cuello de botella
de retroalimentación sin sustituir el juicio docente.
Keywords: turismo sustentable, TBLT, práctica deliberada, inteligencia articial, oralidad
Abstract
is article reported the design and early implementation of a six-week Task-Based Language
Teaching cycle integrated with an AI-assisted “Speaking Studio” for third-semester Sustainable
Tourism students (n=20; 10 hours/week) in Palenque, Mexico, conducted in partnership with
Hotel Misión Palenque. e aim was to improve uency, intelligibility, and the transfer of classroom
English to authentic service encounters. e intervention combined two authentic deliverables—a
90-second concierge pitch and a bilingual micro-artifact—with a weekly routine of deliberate
practice (shadow–record–reect) supported by ChatGPT to generate genre models, phrase banks,
clarity checks, pronunciation targets, and retrieval micro-quizzes. Continuous formative assessment
(exit tickets, error logs) and a summative four-band rubric (intelligibility, uency, appropriacy,
interaction) were planned, complemented by transcript-derived analytics (words per minute, mean
utterance length, lexical variety), partner evaluations, and a short speaking-anxiety scale. Expected
outcomes included faster speech with stable accuracy, improved stress and service-move control,
higher condence when engaging visitors, and the adoption of at least one student-produced artifact
by the hotel. e model proved scalable, low-cost, and replicable, easing the feedback bottleneck
without replacing teacher judgment.
Keywords: sustainable tourism, TBLT, deliberate practice, articial intelligence, speaking
152
Sánchez, C. (Enero – junio 2026). Authentic tourism tasks + ai speaking studio: a scalable model for uency, intelligibility, and transfer. Sathiri, 21
(1), 150 – 154. https://doi.org/10.32645/13906925.1449
Introduction
Tourism undergraduates in public universities often face a misalignment between classroom tasks
and the language demands of real hospitality settings. Teachers also struggle to deliver timely,
individualized speaking feedback at scale. is study addressed both challenges by merging a
community-partner task sequence with an AI-supported practice routine. e article explains the
pedagogical motivation, situates the design in current TESOL discussions on Task-Based Language
Teaching, deliberate practice, and formative assessment, and states the objective: to develop and
evaluate a scalable model that strengthens uency, intelligibility, and transfer to authentic service
encounters at a local hotel.
Objective: to evaluate whether the integrated model improves speaking performance and
perceived condence while remaining feasible within regular course hours.
To clarify two of the key authentic outputs mentioned in this study, it is important to note
that the term ‘90-second concierge pitch’ refers to a brief, realistic oral simulation in which students
assume the role of a hotel concierge to assist a tourist in English. is micro-interaction, limited to
one and a half minutes, aims to develop uency, intelligibility, and service-oriented discourse moves
such as greeting, suggesting, clarifying, and closing the exchange. It provides an ideal benchmark
for measuring intelligibility and condence under realistic time constraints.
Similarly, the term ‘bilingual micro-artifact’ designates a small-scale communicative
product (e.g., signage, service script, or bilingual information card) co-created by students and
hotel sta. Each artifact responds to an authentic need of the partner institution and must be fully
functional in both Spanish and English. It links language learning with real-world impact, allowing
for immediate adoption by the hotel and reinforcing written functional competence in tourism
contexts.
Materials and methods
Design: classroom-based practice with a quasi-experimental, pre/post within-group design.
Participants: n=20 third-semester Sustainable Tourism students (10 hours/week). Setting: University
campus in Palenque, Mexico; community partner Hotel Tren Maya Palenque. Intervention (6
weeks): co-designed tasks culminating in a 90-second concierge pitch and one bilingual micro-
artifact (e.g., signage or script). Weekly ‘Speaking Studio’ routine required students to shadow a
model, record a one-take response, set three micro-goals, and submit transcripts for AI-assisted
feedback. Instruments: four-band speaking rubric; transcript-derived metrics (words per minute,
mean utterance length, type–token proxy); partner evaluation form; short speaking-anxiety
scale; exit tickets and personal error logs. Data analysis: descriptive statistics for all measures and
comparative pre/post tests where applicable; qualitative coding of reections for perceived transfer
and condence.
Results and discussion
At the time of submission, baseline diagnostics and partner briengs had been completed and the
full cycle scheduled. e design is expected to yield measurable gains in uency and intelligibility
and to increase the authenticity of oral output. Discussion centers on how combining TBLT with
AI-supported deliberate practice addresses two persistent constraints: limited contact hours and
feedback bottlenecks. e partner’s role is pivotal for motivation and genre realism, while the AI
153
Authentic Tourism Tasks + AI Speaking
Studio: A Scalable Model for Fluency,
Intelligibility, and Transfer
Sánchez, C. (Enero – junio 2026). Authentic tourism tasks + ai speaking studio: a scalable model for uency, intelligibility, and transfer. Sathiri, 21
(1), 150 – 154. https://doi.org/10.32645/13906925.1449
tool accelerates noticing and repetition without inating teacher workload. Anticipated challenges
include device inequality, over-scripted speech, and potential variability in ASR-based feedback;
mitigations include rehearsal stations, required spontaneous segments, and human-scored rubrics
guiding nal judgments.
Conclusions
An integrated community-partner TBLT cycle with an AI-assisted Speaking Studio oers a practical,
low-cost pathway to improve speaking performance and transfer for Tourism undergraduates. e
model is replicable in similar contexts and aligns with institutional goals of community engagement
and employability. Formal evaluation will document learning gains and feasibility to inform scaling.
Recommendations
Secure a standing memorandum with at least one hotel partner; standardize consent and data-
privacy procedures; time-box AI use to short, regular bursts; and archive high-quality artifacts as
open models for subsequent cohorts. Future research should compare this model against traditional
conversation classes and isolate which components (partner tasks vs. studio drills) drive the largest
gains.
154
Sánchez, C. (Enero – junio 2026). Authentic tourism tasks + ai speaking studio: a scalable model for uency, intelligibility, and transfer. Sathiri, 21
(1), 150 – 154. https://doi.org/10.32645/13906925.1449
References
Derwing, T. M., & Munro, M. J. (2015). Pronunciation fundamentals: Evidence-based perspectives
for L2 teaching. John Benjamins.
Ellis, R. (2009). Task-based language teaching: Sorting out the misunderstandings. International
Journal of Applied Linguistics, 19(3), 221–246.
Long, M. H. (2015). Second language acquisition and task-based language teaching. Wiley-
Blackwell.
Nation, I. S. P. (2013). Learning vocabulary in another language (2nd ed.). Cambridge University
Press.
Schmidt, R. (1990). e role of consciousness in second language learning. Applied Linguistics,
11(2), 129–158.
Wiliam, D. (2011). Embedded formative assessment. Solution Tree Press.
Bygate, M. (2015). Domains and directions in language learning through tasks. John Benjamins.
McCrocklin, S. (2019). ASR-based pronunciation training and its impact on learner attitudes.
In J. Levis (Ed.), Proceedings of the 10th Pronunciation in Second Language Learning &
Teaching Conference (pp. 254–265). Iowa State University.
