- This topic has 1 voice and 0 replies.
-
Posts
-
-
#News(IoTStack) [ via IoTForIndiaGroup ]
Far field voice recognition usually uses microphone array to capture speech from human at a specific distance (In general 1m -10m). And the far field voice recognition system including two parts, front-end signal processing and back-end voice recognition (ASR and NLP). The front-end part aims to use audio enhancement methods and ( including dereverberation, beamforming, etc.), to get a clear audio file (pcm file), the next part is the same as the general voice recognition system.
In this new smart voice command industry, Dusun focus on the front-end voice signal processing algorithms, mainly including Sound Localization(SL), Noise Reduction(NR) , Deliquesce Reverberation (DR), Acoustic Echo Canceller (AEC), Automatic gain control (AGC), as well as Beamforming (BM).
At June 2018, Dusun launches our far-field voice products, voice command modules and microphone array design, to meet needs of numerous industries like IoT devices, home automation, consumer electronics, smart TV, robotics and digital health, and so on. Moreover, Dusun can provide full-stack integration with popular voice engines like Alexa, Google home, Cortana, Baidu and Iflytek, etc.
In the past few years, Dusun has already been specialized in the near-field microphone technology and algorithms, mainly used in voice command remote control for smart TV, and build the long-term relationship with big brands like SKYWORTH, CAAVO, SOMFY, TechniSat, SAMSUNG, etc. With solid experience on voice control technology, Dusun is now capable to offer advanced far-field audio & voice solutions includes a customizable microphone array, a suite of voice algorithms to accurately capture voice commands, integration with popular natural voice engines.
-
- You must be logged in to reply to this topic.
