Smart Home Devices can be used for many different applications such as assisting with daily tasks in the home and office. Many of these applications will rely on far-field voice recognition technology. Therefore, the performance is dictated by the ability of the device to distinguish between the user’s voice, sound played by the device speaker (echo) and ambient noise. Alango is focused on the critical link in this chain - where human meets machine. Alango’s Voice Enhancement Package (VEP) dramatically improves the devices far-field voice pick-up through the implementation of software algorithms that utilize a microphone array to discriminate against noise and home in on the user’s voice.
N-Mic Beamformer | A microphone array utilizing adaptive beamforming preserves the speech from ambient noises, enhances speech from specified directions and helps detecting the direction of voice commands. |
Acoustic Echo Canceller (AEC) | Eliminates acoustic echo from the loudspeaker to the microphones. Both mono and stereo acoustic echo cancellers are available. |
Noise Suppressor (NS) | Detects and attenuates stationary and transient noises. |
Flexible implementation of optimal voice pick up for voice trigger recognition, local (embedded ASR) or transmission to cloud for different environments and number of microphones.
Relatively low computational complexity.
Availability on all major MCU and DSP platforms.
Blind Interference Cancellation (BIC) Technology
|
Direction of Arrival (DOA) Technology
|
Demonstration of a Voice Controlled Smart TV with 6-microphone Linear Array
|
Alango Far-Field Speech Recognition Enhancement Using a Linear Array
|
Alango VEP demo vs. Amazon Echo
|
Sound Processing in Smart Home Devices - the System Approach
|
Far-Field Speech Enhancement from Alango Technologies, Ltd.
|