The young company developing voice AI on the edge with on-device speech recognition and natural language understanding software allows developers to create and fully test voice experiences directly on its recently launched Picovoice Console platform, before exporting models that can run on-device with minimal latency.
Picovoice’s approach to multilingual support is to provide the ability to run multiple speech models simultaneously. Therefore, anyone can create voice experiences that run across platforms and in multi-languages. Picovoice create a demo where users can interact with the voice recognition engine in English, German, French, Italian, Japanese, Korean, Portuguese and Spanish. The demo runs without accessing external services within a web browser. Once the demo is loaded, users can turn off their Internet connection and the demo will keep working!
The company is inviting developers to create an account on its Picovoice Console to train branded wake words and custom voice commands without writing a single line of code. The platform-optimized voice AI models can be downloaded instantly and deployed onto the target platform using the Picovoice SDK in minutes. Currently, the company offers solutions to allow developers to create private, responsive and cost-effective voice products with its Porcupine Wake Word, Rhino Speech-to-Intent, Leopard Speech-to-Text, Octopus Speech-to-Intent and Cobra Voice Activity Detection engines. Its stack can run anywhere from embedded devices to web browsers, resulting in an immersive experience.
www.picovoice.ai
