Providing voice biometry for enhanced security
Speaker ID
DNN based (MFCC feature extraction and threshold-based classification)
Robust performance with scalable options for deployments with Siri, Alexa et all (SDK + Cloud based API)
Providing utterance detection for in car commands so that greater attention is on the driving and hands-free communication
Utterance Detection and STT
Keyword Detection and STT
Pleasanter experience while driving with automated features.
Providing data acquisition services using specific hardware under specific enviromental conditions.
Data Acquisition services
Better datasets leading to better quality of AI product.
Providing active noise cancellation for race drivers
ANC, MVNS, BVS and AEC
MVNS – DNN based mask estimation, BVS – DNN based deep filter estimation, AEC – Soft attention for delay estimation and DNN based complex filtering
Ensuring that drivers and pit crews have better listening experience and thus lesser listening fatigue.
Providing audio clean up package for contents of black box(post every flight/necessity based)
MVNS, BVS, AEC, ANC, ASR (With Speaker Diarisation)
MVNS – DNN based mask estimation, BVS – DNN based deep filter estimation, AEC – Soft attention for delay estimation and DNN based complex filtering
Accurate and reliable audio and transcriptions.