SIGNAL PROCESSING METHOD, SIGNAL PROCESSING DEVICE AND HEARING DEVICE
To provide a technology for physically and acoustically complementing an attenuation vibration characteristic of basement membrane.SOLUTION: When sound data including voice is inputted to an input part 10, a frequency band is divided at a prescribed frequency as a boundary, and the sound data passin...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To provide a technology for physically and acoustically complementing an attenuation vibration characteristic of basement membrane.SOLUTION: When sound data including voice is inputted to an input part 10, a frequency band is divided at a prescribed frequency as a boundary, and the sound data passing through an HPF 12 undergoes FFT. When a vowel fundamental wave extraction part 20 extracts a fundamental wave component or the like of a vowel from a signal of a frequency region obtained by the sound data undergoing the FFT, a synthetic attenuation waveform generation part 22 generates a synthetic attenuation waveform corresponding to a degree of deterioration in frequency selectivity of an individual corresponding to the fundamental wave component or the like of the vowel in a frequency domain, and a synthetic attenuation waveform convolution part 24 sequentially convolves the sound data passing through the HPF 12. An addition part 28 adds the sound data obtained by convolving the synthetic attenuation waveform and undergoing gain adjustment by an adjustment part 26 to the sound data passing through an LPF 14, and output part 30 outputs the sound data. Thus, a spectrum contrast of the voice of frequency band with deteriorated frequency selectivity can substantially be emphasized, and a natural sound with little noise can be outputted.SELECTED DRAWING: Figure 1
【課題】基底膜の減衰振動特性を物理音響的に補完する技術の提供。【解決手段】入力部10に音声を含む音データが入力されると、所定周波数を境界として周波数帯域が分割され、HPF12を通過した音データに対しFFTが施される。これにより得られた周波数領域の信号から、母音基本波抽出部20で母音の基本波成分等が抽出されると、時間領域において、合成減衰波形生成部22で母音の基本波成分等に対応する個人の周波数選択性の劣化度合いに応じた合成減衰波形が生成され、合成減衰波形畳み込み部24でHPF12を通過した音データに逐次畳み込まれる。合成減衰波形が畳み込まれて調整部26で利得の調整がなされた音データは、LPF14を通過した音データと加算部28で加算され、出力部30から出力される。これにより、周波数選択性が劣化した周波数帯域の音声のスペクトルコントラストを実質的に強調でき、異音の少ない自然な音を出力できる。【選択図】図1 |
---|