We provide embedded speech recognition, natural language understanding (NLU) and speech signal processing technologies that are scalable and customizable for on-device and cloud implementations.
Based in Shenzhen, China, VoxAI is a speech recognition and AI software company dedicated to transforming a wide range of consumer products into speech-enabled smart devices. Our Deep Neural Network (DNN) based speech recognition engine is fully scalable – from tiny-footprint, low-power wakeup to full-blown embedded dictation capability – and implemented in a common codebase which we optimize for power and footprint on each device. For applications requiring NLU capabilities, our on-device NLU has you covered – without a network connection. We also offer a complete suite of speech signal processing algorithms such as DNN-based noise suppression – commonly used in hearables – as well as acoustic echo cancelation, beamforming and microphone array solutions for smart speaker-like devices. Every aspect of our software can be customized for each device or platform to achieve optimum performance. Our mission is to remove barriers that prevent consumer products from adopting speech recognition and AI capabilities. No matter what resource constraints you might have on your device or platform, VoxAI can deliver an optimized solution for you.
Established in 2019, VoxAI was founded by world-leading speech recognition experts with over 100 combined years of experience. Our R&D team consists of seasoned acoustic modeling experts, speech recognition decoder experts, speech dialog experts, signal processing experts, embedded architects, and embedded/cloud integration experts. Our sales and marketing team has extensive industry knowledge and deep understanding of the ways to best apply our technology to customer products.