Analytic Wavelet-Packets for Separation of Unknown Speech Signals
نویسندگان
چکیده
Zusammenfassung Die Separation unbekannter Signale ist im Bereich der Sprachsignalverarbeitung von besonderer Bedeutung. Die Trennung erfolgt bei Verwendung mehrerer Mikrofone durch Auswertung der resultierenden Laufzeiten zwischen den einzelnen Quellen und Sensoren. Diese Laufzeitdifferenzen führen im Frequenzbereich zu definierten Phasendifferenzen zwischen den Sensorsignalen. Durch die statistische Analyse der Phasenwerte ist eine Rekonstruktion der Signale unter bestimmten Rahmenbedingungen möglich. Zur Berechnung der spektralen Koeffizienten wird normalerweise die Kurzzeit-Fourier-Transformation (STFT) verwendet. Durch den Einsatz analytischer Wavelet-Packets (AWP) kann die Leistungsfähigkeit derartiger Separationsalgorithmen hinsichtlich Zeitdauer und Genauigkeit verbessert werden. Nach einer Beschreibung des Separationsproblems im Allgemeinen und der Unterschiede der beiden Zeit-Frequenz-Darstellungen werden die Vorteile der Wavelet-Packets gegenüber der KurzzeitFourier-Transformation erläutert. Summary The separation of a mixture of unknown speech signals is a demanding research topic in the area of signal processing. Recording the signals with two sensors at different positions, the resulting time delay of arrival can be used to reconstruct the original signals. In the frequency domain the delay is related to characteristic phase differences between the sensor signals. Under certain assumptions the original signals can be reconstructed based on a statistical analysis of the phase values. To this end, the signals have to be transformed to the time-frequency domain due to the non-stationary character of speech signals. Usually, the short-time Fourier transform (STFT) is used to calculate the spectral coefficients. Using analytic wavelet packets (AWP) the performance of the separation algorithm with regard to both computation time and accuracy can be improved.
منابع مشابه
A New Algorithm for Voice Activity Detection Based on Wavelet Packets (RESEARCH NOTE)
Speech constitutes much of the communicated information; most other perceived audio signals do not carry nearly as much information. Indeed, much of the non-speech signals maybe classified as ‘noise’ in human communication. The process of separating conversational speech and noise is termed voice activity detection (VAD). This paper describes a new approach to VAD which is based on the Wavelet ...
متن کاملAn Adaptive Algorithm for Speech Source Separation in Overcomplete Cases Using Wavelet Packets
Speech process has benefited a great deal from the wavelet transforms. Wavelet packets decompose signals in to broader components using linear spectral bisecting. In this paper, mixtures of speech signals are decomposed using wavelet packets, the phase difference between the two mixtures are investigated in wavelet domain. A geometrical model is defined and an adaptive algorithm is proposed bas...
متن کاملUnderdetermined Blind Mixing Matrix Estimation Using STWP Analysis for Speech Source Signals
Wavelet packets decompose signals in to broader components using linear spectral bisecting. Mixing matrix is the key issue in the Blind Source Separation (BSS) literature especially in under-determined cases. In this paper, we propose a simple and novel method in Short Time Wavelet Packet (STWP) analysis to estimate blindly the mixing matrix of speech signals from noise free linear mixtures in ...
متن کاملBlind Source Separation via Multinode Sparse Representation
We consider a problem of blind source separation from a set of instantaneous linear mixtures, where the mixing matrix is unknown. It was discovered recently, that exploiting the sparsity of sources in an appropriate representation according to some signal dictionary, dramatically improves the quality of separation. In this work we use the property of multi scale transforms, such as wavelet or w...
متن کاملA New Method for Speech Enhancement Based on Incoherent Model Learning in Wavelet Transform Domain
Quality of speech signal significantly reduces in the presence of environmental noise signals and leads to the imperfect performance of hearing aid devices, automatic speech recognition systems, and mobile phones. In this paper, the single channel speech enhancement of the corrupted signals by the additive noise signals is considered. A dictionary-based algorithm is proposed to train the speech...
متن کامل