mlsanityData to 7 Oct 2026

LuxSQA: Ask Me in Luxembourgish with TTS-Augmented Spoken Question Answering

arXiv 2607.02763 ↗First posted 2 Jul 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Spoken Question Answering (SQA) remains largely focused on high-resource languages and carefully recorded speech, limiting the reach of speech-LLM methods in low-resource settings. This paper investigates whether text-to-speech (TTS) can provide task-specific training data for Luxembourgish SQA without requiring a large human-recorded QA corpus. Starting from existing text-based QA resources, we translate questions into Luxembourgish, synthesize spoken questions with multiple TTS systems, and pair them with textual answers. We train a parameter-efficient SLAM-style architecture that connects a frozen Whisper encoder to frozen multilingual LLM backends through a learned projector and LoRA adapters. We compare MMS-TTS, Qwen3-TTS, and OmniVoice variants, including single-source corpora of about 48k questions and a 4TTS multi-source mix of approximately 230k questions. Evaluation on LLAMA-LB-Test with two real Luxembourgish speaker conditions shows that multi-source and voice-design-based synthetic training configurations yield the strongest SQA performance. The results also show that no-reference TTS quality scores do not monotonically predict downstream QA performance, indicating that synthetic speech must be evaluated as task-specific training data rather than only as natural-sounding audio.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 32 references that are in the map. First posted on arXiv 2 Jul 2026.

Builds on

26 of 32 references.

References · 26

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  2. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  3. 2023Scaling Speech Technology to 1,000+ Languagescited by 401
  4. 2020DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressorscited by 235
  5. 2026Qwen3-TTS Technical Reportcited by 106
  6. 2024EuroLLM: Multilingual Language Models for Europecited by 103
  7. 2023Spoken Question Answering and Speech Continuation Using Spectrogram-Powered LLMcited by 92
  8. 2024An Embarrassingly Simple Approach for LLM with Strong ASR Capacitycited by 86
  9. 2023SLM: Bridge the thin gap between speech and text foundation modelscited by 67
  10. 2024SLAM-Omni: Timbre-Controllable Voice Interaction System with Single-Stage Trainingcited by 58
  11. 2022End-to-end Spoken Conversational Question Answering: Task, Dataset and Modelcited by 37
  12. 2021Deep Learning Based Assessment of Synthetic Speech Naturalnesscited by 35
14 more
  1. 2020Towards Data Distillation for End-to-end Spoken Conversational Question Answeringcited by 34
  2. 2021Self-supervised Dialogue Learning for Spoken Conversational Question Answeringcited by 27
  3. 2026OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Modelscited by 22
  4. 2024Performance evaluation of SLAM-ASR: The Good, the Bad, the Ugly, and the Way Forwardcited by 17
  5. 2023HeySQuAD: A Spoken Question Answering Datasetcited by 16
  6. 2023LibriSQA: A Novel Dataset and Framework for Spoken Question Answering with Large Language Modelscited by 13
  7. 2024SpeechDPR: End-to-End Spoken Passage Retrieval for Open-Domain Spoken Question Answeringcited by 12
  8. 2025InSerter: Speech Instruction Following with Unsupervised Interleaved Pre-trainingcited by 11
  9. 2022Data Augmentation for Low-Resource Quechua ASR Improvementcited by 8
  10. 2023GSQA: An End-to-End Model for Generative Spoken Question Answeringcited by 8
  11. 2025Speech LLMs in Low-Resource Scenarios: Data Volume Requirements and the Impact of Pretraining on High-Resource Languagescited by 7
  12. 2026SLAM-LLM: A Modular, Open-Source Multimodal Large Language Model Framework and Best Practice for Speech, Language, Audio and Music Processingcited by 7
  13. 2022On the Impact of Speech Recognition Errors in Passage Retrieval for Spoken Question Answeringcited by 6
  14. 2024On the Problem of Text-To-Speech Model Selection for Synthetic Data Generation in Automatic Speech Recognitioncited by 3

Built on by

Not cited yet.