公开(公告)号：US10993046B2

公开(公告)日：2021-04-27

申请号：US16462145

申请日：2017-08-25

Applicant: POSTECH ACADEMY-INDUSTRY FOUNDATION

Inventor： Hong June Park ,  Seong Eun Cho ,  Ki Hwan Sung ,  Won Cheol Lee ,  Hyeon Kyu Noh ,  Im Soo Mok

IPC: H04R25/00 ,  H04M1/725 ,  H04R3/00 ,  H04M1/72409 ,  G06F5/06 ,  G06F13/10 ,  G06F13/38

Abstract: The present invention relates to a smartphone-based hearing aid which does not employ a separate power supply or digital signal processing device and instead has an analogue interface unit, configured with a single integrated circuit chip, connected between a plurality of microphones and a smartphone, thereby realizing a compact size, reducing the cost and enhancing the audio performance. Instead of employing a digital signal processing chip which is essentially used in the existing hearing aids or personal sound amplifiers, the smartphone-based hearing aid according to the present invention utilizes components such as a smartphone-embedded application processor (AP), RAM, digital-analogue converter, speaker and display and is therefore affordable. In particular, if a CPU and a GPU which are embedded in the application processor of the smartphone are utilized, digital signal processing necessary for the operation of the hearing aid can be performed within a short time of milliseconds (ms) just by using software.

公开(公告)号：US11183177B2

公开(公告)日：2021-11-23

申请号：US16626862

申请日：2018-04-26

Applicant: POSTECH ACADEMY-INDUSTRY FOUNDATION

Inventor： Hong June Park ,  Hyeon Kyu Noh ,  Won Cheol Lee ,  Kyeong Won Jeong

IPC: G10L15/187 ,  G06F21/76 ,  G06N3/08 ,  G10L15/02 ,  G10L15/16 ,  G10L15/22 ,  G10L13/00

Abstract: The present invention relates to a real-time voice recognition apparatus equipped with an application-specific integrated circuits (ASIC) chip and a smartphone, capable, by using one smartphone and one ASIC chip and without using a cloud computer, of assuring personal privacy, and, due to a short delay time, enabling real-time conversion of voice input signals into text for output. When one DRAM chip is optionally added to the real-time voice recognition apparatus, the number of neural network layers is increased thereby significantly improving accuracy of conversion of voice input signals into text.