• MALSORI
• /
• no.47
• /
• pp.97-107
• /
• 2003

Speaker change detection is to detect automatically a point of time at which speaker was replaced. Since feature parameters used for speaker change detection depend not only on speaker characteristics but also on phonetic characteristics, spoken contents included in the feature parameters inevitably causes performance degradation of speaker change detection. In this paper, to alleviate this problem, a method to normalize phonetic variations in speech feature parameters is proposed for emphasizing changes due to speaker characteristics. Experimental results show that the proposed method improves the performance of speaker change detection.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.466-472
• /
• 2021

Speaker Change Detection (SCD) refers to finding the moment when the main speaker changes from one person to the next in a speech conversation. In speaker change detection, difficulties arise due to overlapping speakers, inaccuracy in the information labeling, and data imbalance. To solve these problems, TIMIT corpus widely used in speech recognition have been concatenated artificially to obtain a sufficient amount of training data, and the detection of changing speaker has performed after identifying overlapping speakers. In this paper, we propose an speaker change detection system that considers the speaker overlapping. We evaluated and verified the performance using various approaches. As a result, a detection system similar to the X-Vector structure was proposed to remove the speaker overlapping region, while the Bi-LSTM method was selected to model the speaker change system. The experimental results show a relative performance improvement of 4.6 % and 13.8 % respectively, compared to the baseline system. Additionally, we determined that a robust speaker change detection system can be built by conducting related studies based on the experimental results, taking into consideration text and speaker information.

• Speech Sciences
• /
• v.8 no.4
• /
• pp.59-72
• /
• 2001

In this paper, we experimented with speaker change detection that uses a statistical method for NOD (News On Demand) service. A specified speaker's change can find out content of each data in speech if analysed because it means change of data contents in news data. Speaker change detection acts as preprocessor that divide input speech by speaker. This is an important preprocessor phase for speaker tracking. We detected speaker change using GLR(generalized likelihood ratio) distance base division and BIC (Bayesian information criterion) base division among matrix method. An experiment verified speaker change point using BIC base division after divide by speaker unit using GLR distance base method first. In the experimental result, FAR (False Alarm Rate) was 63.29 in high noise environment and FAR was 54.28 in low noise environment in MDR (Missed Detection Rate) 15% neighborhood.

• The Journal of the Acoustical Society of Korea
• /
• v.30 no.2
• /
• pp.80-85
• /
• 2011

Speaker change detection involves the identification of time indices of an audio stream, where the identity of the speaker changes. In this paper, we propose novel measures for the speaker change detection based on a graph-partitioning criterion over the pairwise distance matrix of feature-vector stream. Experiments on both synthetic and real-world data were performed and showed that the proposed approach yield promising results compared with the conventional statistical measures.

• Journal of Internet Computing and Services
• /
• v.11 no.6
• /
• pp.21-29
• /
• 2010

In this paper, we propose the system which detects the speaker, who is speaking currently, among a number of users. A proposed speaker detection system based on stereo vision and audio is mainly composed of the followings: a position estimation of speaker candidates using stereo camara and microphone, a current speaker detection, and a speaker information acquisition based on a mobile device. We use the haar-like features and the adaboost algorithm to detect the faces of speaker candidates with stereo camera, and the position of speaker candidates is estimated by a triangulation method. Next, the Time Delay Of Arrival (TDOA) is estimated by the Cross Power Spectrum Phase (CPSP) analysis to find the direction of source with two microphone. Finally we acquire the information of the speaker including his position, voice, and face by comparing the information of the stereo camera with that of two microphone. Furthermore, the proposed system includes a TCP client/server connection method for mobile service.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.7
• /
• pp.69-75
• /
• 2001

Intra-speaker variation, noise environments, and mismatch between training and test conditions are the major reasons for the speaker verification system unable to use it practically. In this study, we propose robust end-point detection algorithm, noise cancelling with the microphone property compensation technique, and inter-speaker discriminate technique by weighting cepstrum for robust speaker verification system. Simulation results show that the average speaker verification rate is improved in the rate of 17.65% with proposed end-point detection algorithm using LPC residue and is improved in the rate of 36.93% with proposed noise cancelling and microphone property compensation algorithm. The proposed weighting function for discriminating inter-speaker variations also improves the average speaker verification rate in the rate of 6.515%.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.17 no.5
• /
• pp.68-79
• /
• 2003

In this paper, we propose a system that detects the current speaker in multi-speaker video conference by using lip motion. First, the system detects the face and lip area of each of the speakers using face color and shape information. Then, to detect the current speaker, it calculates the change between the current frame and the previous frame. To accomplish this, we used two CCD cameras. One is a general CCD camera, the other is a PTZ camera controlled by RS-232C serial port. The result is a system capable of detecting the face of current speaker in a video feed with more than three people, regardless of orientation of the faces. With this system, it only takes 4 to 5 seconds to zoom in on the speaker from the initial image. Also, it is amore efficient image transmission system for such things as video conference and internet broadcasting because it offers a face area screen at a resolution of 320X240, while at the same time providing a whole background screen.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.137-140
• /
• 2003

In this paper, we propose a robust endpoint detection algorithm for speaker verification. Proposed algorithm uses energy and cepstral distance parameters, and it replaces the detected endpoints with endpoints of voiced speech, when the estimated signal-to-noise ratio (SNR) is low. Experimental results show that proposed algorithm is superior to energy-based endpoint detection algorithm.

• Speech Sciences
• /
• v.12 no.3
• /
• pp.125-136
• /
• 2005

Automatic speech recognition is much more difficult in real world. Speech recognition according to SIR (Signal to Interface Ratio) is difficult in situations in which noise of surrounding environment and multi-speaker exists. Therefore, study on main speaker's voice extractions a very important field in speech signal processing in binaural sound. In this paper, we used directional filter and harmonic filter among other existing methods to extract the main speaker's information in binaural sound. The main speaker's voice was extracted using directional filter, and other remaining speaker's information was removed using harmonic filter through main speaker's pitch detection. As a result, voice of the main speaker was enhanced.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.4
• /
• pp.266-272
• /
• 2012

Speaker segmentation is a process of automatically detecting the speaker boundary points in the audio data. Speaker segmentation methods are divided into two categories depending on whether they use a prior knowledge or not: One is the model-based segmentation and the other is the metric-based segmentation. In this paper, we introduce the eigenvoice-based speaker weight distance method and compare it with the representative metric-based methods. Also, we employ and compare the Euclidean and cosine similarity functions to calculate the distance between speaker weight vectors. And we verify that the speaker weight distance method is computationally very efficient compared with the method directly using the distance between the speaker adapted models constructed by the eigenvoice technique.