서 론
![]() | Fig. 1.Representative images showing the decomposition of original voice by using the Hilbert transform into two forms of temporal information, temporal envelope and TFS. Time waveform of an original voice (A). The envelope obtained from the original voice shows slow fluctuation over time (B). The combination of the envelope and TFS. Note that the original voice can be derived by modulating the TFS signal with the envelope (C). The TFS obtained from the original voice shows no fluctuation over time (D). TFS: temporal fine structure. |

본 론
음성 인지에서 포락선 정보와 미세구조 정보의 분석 방법
![]() | Fig. 2.Example of how to form a chimera wave. Two different band-limited input signals are decomposed into their envelope and fine structure using the Hilbert transform. Then, the envelope 1 and the temporal fine structure 2 are synthesized to form a single-band auditory chimera. These partial chimeras are finally summed over all frequency bands to produce a multi-band chimera. |
소음 환경에서 음성 인식과 시간적 미세구조 정보의 역할
난청 환자에서의 시간적 미세구조 정보의 인지
인공 와우와 시간적 미세구조 정보
![]() | Fig. 6.Illustration of the harmonic electrode matching procedure. The Harmonic-Single-Sideband-Encoder tracks the harmonics of a single musical source and transforms them into modulators conveying both amplitude and temporal fine structure cues to electrodes. The procedure starts with the identification of stronger harmonics and assigns them to appropriate electrodes. This harmonic electrode matching procedure was executed repeatedly until only 8 stronger harmonics were selected for electric stimulation. Throughout the duration of a note, the relative magnitudes between harmonics may change and this relative timing of their onsets and offsets can be adequately represented for timbre perception. |
![]() | Fig. 7.Illustration of fine structure processing strategy. This strategy uses the timing of stimulation to code the temporal fine structure of the sound signal in the low frequencies. The time coding is achieved using CSSS. CSSS are series of stimulation pulses which are triggered by zero-crossing in a channel’s bandpass filter output. CSSS: Channel-Specific Sampling Sequences. |
기 타

결 론
