DIGITAL RADIO TRANSMITTER AND DIGITAL RADIO COMMUNICATION SYSTEM

• Main Authors: TAMADA SHIGEHISA, VO MINH NHUT, KOKUBU JIRO, HIDAKA KOJI
• Format: Patent
• Language: English; Japanese
• Published: 12.05.2023
• Subjects: ACOUSTICS; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

To remove chime sound which becomes abnormal sound in an individual house station and to output only a voice signal without the need to perform complicated signal processing such as "generation of a non-voice code after sound source pattern analysis processing of previously recorded non-voice data and a non-voice signal by AD conversion is performed".SOLUTION: A digital radio transmitter digitally modulates a low frequency signal including a voice signal and a non-voice signal by a digital modulator by using a prescribed voice compression system and radio-transmits the signal. The digital radio transmitter includes a noise cancel unit for performing voice signal processing so as to cancel noise from the low frequency signal including the voice signal and the non-voice signal and outputting the signal to the digital modulator. The noise cancel unit includes a voice signal processing unit which trains a deep learning model unit for determining whether or not it is a non-voice period including noise from the inputted voice signal by using a feature parameter of human voice, and includes the learned deep learning model unit.SELECTED DRAWING: Figure 1 【課題】「予め記録した非音声データとＡＤ変換による非音声信号の音源パターン解析処理を行ってから非音声コードを生成」などの複雑な信号処理を行う必要がなく、戸別局において異音となったチャイム音を除去して、音声信号のみを出力する。【解決手段】デジタル無線送信装置は、音声信号及び非音声信号を含む低周波信号を、デジタル変調器により所定の音声圧縮方式を用いてデジタル変調して無線送信する。デジタル無線送信装置は、音声信号及び非音声信号を含む低周波信号からノイズをキャンセルするように音声信号処理を行ってデジタル変調器に出力するノイズキャンセル部を備える。ノイズキャンセル部は、人間の音声の特徴パラメータを用いて、入力される音声信号からノイズを含む非音声期間であるか否かを判定する深層学習モデル部を学習して、学習された深層学習モデル部を構成して備える音声信号処理部を備える。【選択図】図１