SMCS AI Lab

Based at Massey University, the SMCS AI Lab focuses on building leading real-world applications across diverse domains using artificial intelligence. Our research is supported by multiple large-scale projects funded by the New Zealand government. We develop advanced AI systems for speech recognition, speech synthesis, computer vision, image processing, and sign language understanding, with a strong emphasis on robustness, scalability, and real-world deployment.

RESEARCH THEMES

  • • Automatic Speech Recognition
  • • Speech Synthesis
  • • Natural Language Processing
  • • Computer Vision & Image Processing
  • • Video Caption
  • • Sign Language Recognition & Generation
  • • Multimodal and Conversational AI
Lab photo

Our Projects

Māori Coversation

消费类电子

We developed an end-to-end Māori conversational system that supports natural language interaction through both speech and text. The system integrates automatic speech recognition, natural language understanding, dialogue management, and speech synthesis, enabling users to engage in fluent conversations in te reo Māori……

+ More Details

Sign Language

移动通讯电子

A deep learning–based system for both sign language recognition and sign language generation, designed to model the complex spatiotemporal and linguistic properties of sign languages. For sign language recognition, the system employs spatiotemporal neural architectures, including convolutional and transformer-based models ...

+ More Details

Speech Synthesis

移动通讯电子

Māori Speech Synthesis System is an advanced text-to-speech (TTS) platform designed to support the revitalization and digital inclusion of te reo Māori. The system converts written Māori text into natural and intelligible speech using state-of-the-art deep learning models trained on carefully curated Māori speech data. By capturing the phonetic characteristics, prosody. ...

+ More Details
 

Datasets

Name Description Type Version Action
Multilingual Sign Languages This repository documents multilingual sign language benchmark datasets for researchers. Our goal is to support multilingual, cross-cultural sign language understanding and generation by leveraging diverse large-scale datasets. Sign Languages v1 Details
Punjabi Speech The Punjabi Speech corpus is designed for automatic speech recognition and synthesis purposes. Speech v2 Details
Address:Massey University East Precinct Albany Expressway, SH17, Albany, Auckland 0632 Contact:64-022-0889176