Developing multilingual speech synthesis system for Ojibwe, Mi'kmaq, and Maliseet

2025/02/04 by Wang, Shenran, Yang, Changbing, Parkhill, Mike +3 · 1 citation
#Artificial Intelligence (cs.AI) #Audio and Speech Processing (eess.AS) #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2502.02703

Abstract

We present lightweight flow matching multilingual text-to-speech (TTS) systems for Ojibwe, Mi'kmaq, and Maliseet, three Indigenous languages in North America. Our results show that training a multilingual TTS model on three typologically similar languages can improve the performance over monolingual models, especially when data are scarce. Attention-free architectures are highly competitive with self-attention architecture with higher memory efficiency. Our research not only advances technical development for the revitalization of low-resource languages but also highlights the cultural gap in human evaluation protocols, calling for a more community-centered approach to human evaluation.

Cited by

• LingGym: How Far Are LLMs from Thinking Like Field Linguists?

Related

• Automatic Speech Recognition Advancements for Indigenous Languages of the Americas
• Redefining technology for indigenous languages
• Is It Navajo? Accurate Language Detection in Endangered Athabaskan Languages
• Challenges of language technologies for the indigenous languages of the Americas
• Enhancing Translation for Indigenous Languages: Experiments with Multilingual Models