Use Case:
Voice Recognition for Kiosks & POS
Description
Self-Service Kiosks at retail, showrooms, restaurants, hospitality, office buildings and event venues to handle customer self-ordering, self-check in/out, self-registration, etc. by using voice interaction to enhance the user experiences and the efficiency of transactions. The kiosks need to hear and understand the customer’s voice commands in high noise environments and often with other people talking close-by.
Challenges
- Noises in the background
- Noises in the store, mall, restaurant, airport, etc., can prevent a kiosk from understanding a customer’s voice commands
- Competing voice from users of other near-by kiosks can lead to misunderstanding of the voice command
- Self-service kiosks are often located in large spaces that cause sound to bounce back and forth, resulting in strong reverberation that interfere with clear voice being heard by the kiosks
- Customers can be annoyed waiting for Kiosk’s voice prompts or audio playback contents to finish
Solution Proposed
Ark X AFE:
- Audio Front End supports
- Multi-mic noise reduction removes stationary and transient background noise, resulting in accurate voice recognition even in noisy conditions
- AFE uses multi-mic algorithm and exploits reverberant propagation to identify the primary talker that is using the kiosk and reject other near-by talkers
- De-reverberation processing enhances the intelligibility of speech in large spaces
- Excellent echo cancellation enables customer to barge-in and interrupt a kiosk’s voice prompts and audio playback at anytime
Far-field Voice Applications
Our voice capture technology can easily be applied to a wide range of voice-enabled smart products and devices.