Maleo Research Group

We are a non-profit research group dedicated to advancing open-source artificial intelligence, with a specialized focus on Speech and Audio Processing. Our core research interests lie in pushing the boundaries of Speaker Diarization and Automatic Speech Recognition (ASR) to create more robust, context-aware audio models.

🎯 Research Interests

• Speaker Diarization: Segmenting and identifying "who spoke when" in complex, multi-speaker environments.
• Speech Recognition: Developing highly accurate ASR systems, particularly for underrepresented contexts, emotional speech, and noisy backgrounds.
• Audio Classification & Audio Scene Analysis: Understanding the environment and emotional context surrounding speech.

🤖 Models

Our models focus on real-world audio challenges, including noise classification and environmental context.

📊 Datasets

We curate high-quality audio datasets to support speech recognition, emotion detection, and audio analysis research.

🤝 Connect & Collaborate

As a non-profit research collective, we highly value collaboration, open science, and community feedback.

• Contribute: If you find our models or datasets useful, feel free to open an issue or discussion on our repositories.
• Contact: Reach out to us through our Hugging Face Organization page for research collaborations or inquiries.