Journal ArticleDOI
Separation of speech from interfering speech by means of harmonic selection
Reads0
Chats0
TLDR
In this paper, the harmonics of the desired voice in the Fourier transform of the input were selected to distinguish between two different voices. But the authors focus on the principal subproblem, the separation of vocalic speech.Abstract:
A common type of interference in speech transmission is that caused by the speech of a competing talker. Although the brain is adept at clarifying such speech, it relies heavily on binaural data. When voices interfere over a single channel, separation is much more difficult and intelligibility suffers. Clarifying such speech is a complex and varied problem whose nature changes with the moment‐to‐moment variation in the types of sound which interfere. This paper describes an attack on the principal subproblem, the separation of vocalic speech. Separation is done by selecting the harmonics of the desired voice in the Fourier transform of the input. In implementing this process, techniques have been developed for resolving overlapping spectrum components, for determining pitches of both talkers, and for assuring consistent separation. These techniques are described, their performance on test utterances is summarized, and the possibility of using this process as a basis for the solution of the general two‐tal...read more
Citations
More filters
Análisis y síntesis de señales de audio a través de la Transformada Wavelet continua y compleja: El algoritmo CWAS
TL;DR: In this paper, the Transformada Wavelet Continua and Compleja (CCWT) is proposed as a herramienta precisa for the obtencion of caracteristicas of alto nivel of the senal de audio, a traves of un algoritmo generalista del modelo that se propone de the samea.
Proceedings ArticleDOI
Single channel speech separation based on sinusoidal modeling
TL;DR: This work presents an approach of single channel speech separation based on sinusoidal modeling and multi-scale product analysis, which effectively estimates multi-pitch of the desired speaker in a multiple speaker context.
Proceedings ArticleDOI
A voice separation system based on phoneme segregation
TL;DR: A new voice separation system for improving the intelligibility of speech that has been interfered by another talker's voice is presented, based on a sinusiodal model of phoneme and a phoneme segregation algorithm.
Journal ArticleDOI
A Harmonics Extraction Filter with Robustness to Pitch and Amplitude Fluctuation
Kazuki Nishi,Shigeru Ando +1 more
TL;DR: In this article, an optimal harmonics extraction filter with pitch fluctuation from the noise-corrupted observation is derived as a solution to the Kalman-Bucy filter in which the pitch fluctuations are treated as a state-dependent noise.