This repository is a curated list of research papers, open-source code, software, and datasets related to Speech Keyword Spotting (KWS), also known as Wake-Up Word Detection. It serves as a valuable resource for researchers and developers working on hands-free voice control for IoT devices, aiming for low-power, low-latency, and accurate keyword detection.

How It Works

The project curates a comprehensive collection of academic publications, categorizing them by year and highlighting key contributions. It also lists relevant open-source code repositories and software toolkits that implement KWS techniques, alongside datasets specifically designed for training and evaluating KWS systems. This structured approach allows users to quickly find state-of-the-art methods and practical implementations.

Quick Start & Requirements

This repository is a curated list and does not have a direct installation or execution command. Users are directed to individual linked repositories for specific setup instructions.

Highlighted Details

• Extensive collection of publications from 2014 to present, covering various KWS techniques.
• Links to numerous open-source code repositories implementing KWS, including those with state-of-the-art results.
• Curated list of datasets like Speech Commands, Mobvoi Hotwords, and HI-MIA for KWS model training and evaluation.
• Mentions of software toolkits like WeKws, Porcupine, Sonus, Picovoice, and Mycroft Precise for practical KWS implementation.

Maintenance & Community

The repository is maintained by zycv. Specific community links (e.g., Discord, Slack) or active development signals are not explicitly provided in the README.

Licensing & Compatibility

The repository itself is a list and does not impose a license. However, the linked open-source code and software projects will have their own licenses, which users must consult for compatibility and usage restrictions.

Limitations & Caveats

This is a curated list, not a functional software project. Users must navigate to individual linked resources to find executable code, datasets, or software, and are responsible for understanding the requirements and licenses of those external projects.