TaigiSpeech: A Low-Resource Real-World Speech Intent Dataset and Preliminary Results with Scalable Data Mining In-the-Wild

• URL: http://arxiv.org/abs/2603.21478v1
• Date: Mon, 23 Mar 2026 01:44:45 GMT
• Title: TaigiSpeech: A Low-Resource Real-World Speech Intent Dataset and Preliminary Results with Scalable Data Mining In-the-Wild
• Abstract summary: Speech technologies have advanced rapidly and serve diverse populations worldwide.<n>Many languages remain underrepresented due to limited resources.<n>We introduce textbfTaigiSpeech, a real-world speech intent dataset in Taiwanese Taigi.
• Score: 102.11425887660327
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Speech technologies have advanced rapidly and serve diverse populations worldwide. However, many languages remain underrepresented due to limited resources. In this paper, we introduce \textbf{TaigiSpeech}, a real-world speech intent dataset in Taiwanese Taigi (aka Taiwanese Hokkien/Southern Min), which is a low-resource and primarily spoken language. The dataset is collected from older adults, comprising 21 speakers with a total of 3k utterances. It is designed for practical intent detection scenarios, including healthcare and home assistant applications. To address the scarcity of labeled data, we explore two data mining strategies with two levels of supervision: keyword match data mining with LLM pseudo labeling via an intermediate language and an audio-visual framework that leverages multimodal cues with minimal textual supervision. This design enables scalable dataset construction for low-resource and unwritten spoken languages. TaigiSpeech will be released under the CC BY 4.0 license to facilitate broad adoption and research on low-resource and unwritten languages. The project website and the dataset can be found on https://kwchang.org/taigispeech.

Related papers

• WAXAL: A Large-Scale Multilingual African Language Speech Corpus [12.433885475371035]
  WAXAL is a large-scale, openly accessible speech dataset for 21 languages representing over 100 million speakers.<n>The collection consists of two main components: an Automated Speech Recognition (ASR) dataset containing approximately 1,250 hours of transcribed, natural speech from a diverse range of speakers, and a Text-to-Speech (TTS) dataset with over 180 hours of high-quality, single-speaker recordings reading phonetically balanced scripts.
  arXiv  Detail & Related papers  (2026-02-02T19:49:19Z)
• A multilingual training strategy for low resource Text to Speech [5.109810774427171]
  We investigate whether data from social media can be used for a small TTS dataset construction, and whether cross lingual transfer learning can work with this type of data. To do so, we explore how data from foreign languages may be selected and pooled to train a TTS model for a target low resource language. Our findings show that multilingual pre-training is better than monolingual pre-training at increasing the intelligibility and naturalness of the generated speech.
  arXiv  Detail & Related papers  (2024-09-02T12:53:01Z)
• Learning to Speak from Text: Zero-Shot Multilingual Text-to-Speech with Unsupervised Text Pretraining [65.30528567491984]
  This paper proposes a method for zero-shot multilingual TTS using text-only data for the target language. The use of text-only data allows the development of TTS systems for low-resource languages. Evaluation results demonstrate highly intelligible zero-shot TTS with a character error rate of less than 12% for an unseen language.
  arXiv  Detail & Related papers  (2023-01-30T00:53:50Z)
• Speech-to-Speech Translation For A Real-world Unwritten Language [62.414304258701804]
  We study speech-to-speech translation (S2ST) that translates speech from one language into another language. We present an end-to-end solution from training data collection, modeling choices to benchmark dataset release.
  arXiv  Detail & Related papers  (2022-11-11T20:21:38Z)
• Virtuoso: Massive Multilingual Speech-Text Joint Semi-Supervised Learning for Text-To-Speech [37.942466944970704]
  This paper proposes Virtuoso, a massively multilingual speech-text joint semi-supervised learning framework for text-to-speech synthesis (TTS) models. To train a TTS model from various types of speech and text data, different training schemes are designed to handle supervised (TTS and ASR data) and unsupervised (untranscribed speech and unspoken text) datasets. Experimental evaluation shows that multilingual TTS models trained on Virtuoso can achieve significantly better naturalness and intelligibility than baseline ones in seen languages.
  arXiv  Detail & Related papers  (2022-10-27T14:09:48Z)
• ASR2K: Speech Recognition for Around 2000 Languages without Audio [100.41158814934802]
  We present a speech recognition pipeline that does not require any audio for the target language. Our pipeline consists of three components: acoustic, pronunciation, and language models. We build speech recognition for 1909 languages by combining it with Crubadan: a large endangered languages n-gram database.
  arXiv  Detail & Related papers  (2022-09-06T22:48:29Z)
• Automatic Speech Recognition Datasets in Cantonese Language: A Survey and a New Dataset [85.52036362232688]
  Our dataset consists of 73.6 hours of clean read speech paired with transcripts, collected from Cantonese audiobooks from Hong Kong. It combines philosophy, politics, education, culture, lifestyle and family domains, covering a wide range of topics. We create a powerful and robust Cantonese ASR model by applying multi-dataset learning on MDCC and Common Voice zh-HK.
  arXiv  Detail & Related papers  (2022-01-07T12:09:15Z)
• Consecutive Decoding for Speech-to-text Translation [51.155661276936044]
  COnSecutive Transcription and Translation (COSTT) is an integral approach for speech-to-text translation. The key idea is to generate source transcript and target translation text with a single decoder. Our method is verified on three mainstream datasets.
  arXiv  Detail & Related papers  (2020-09-21T10:10:45Z)
• CoVoST 2 and Massively Multilingual Speech-to-Text Translation [24.904548615918355]
  CoVoST 2 is a large-scale multilingual speech translation corpus covering translations from 21 languages into English and from English into 15 languages. This represents the largest open dataset available to date from total volume and language coverage perspective.
  arXiv  Detail & Related papers  (2020-07-20T17:53:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.