Speaker Identification with VoxCeleb DataSet
نویسندگان
چکیده
In this project, we perform a text independent speaker identification experiment with a newly released data set, VoxCeleb (2017)[1], which consists of celebrity interview audio clips downloaded from Youtube. It’s a challenging data set in the sense that there are often multiple vocal sources in the same clip. A MFCC feature vector based Deep Neural Network (DNN) is used as our baseline. It is compared with a support vector machine (SVM) algorithm. We utilzes the k-mean vector quantization technique to optimize our SVM performance, as well as filter out other vocal sources while training the SVM model.
منابع مشابه
VoxCeleb: A Large-Scale Speaker Identification Dataset
Most existing datasets for speaker identification contain samples obtained under quite constrained conditions, and are usually hand-annotated, hence limited in size. The goal of this paper is to generate a large scale text-independent speaker identification dataset collected ‘in the wild’. We make two contributions. First, we propose a fully automated pipeline based on computer vision technique...
متن کاملAttentive Statistics Pooling for Deep Speaker Embedding
This paper proposes attentive statistics pooling for deep speaker embedding in text-independent speaker verification. In conventional speaker embedding, frame-level features are averaged over all the frames of a single utterance to form an utterance-level feature. Our method utilizes an attention mechanism to give different weights to different frames and generates not only weighted means but a...
متن کاملPhonetic Speaker Id
This paper describes the exploration of text-independent speaker identification using novel approaches based on speakers’ phonetic features instead of traditional acoustic features. Different phonetic speaker identification approaches are discussed in this paper and evaluated using two speaker identification systems: one multilingual system and one single language multiple-engine system. Furthe...
متن کاملSpeaker Identification for Swiss German with Spectral and Rhythm Features
We present results of speech rhythm analysis for automatic speaker identification. We expand previous experiments using similar methods for language identification. Features describing the rhythmic properties of salient changes in signal components are extracted and used in an speaker identification task to determine to which extent they are descriptive of speaker variability. We also test the ...
متن کاملSeeing Voices and Hearing Faces: Cross-modal biometric matching
We introduce a seemingly impossible task: given only an audio clip of someone speaking, decide which of two face images is the speaker. In this paper we study this, and a number of related cross-modal tasks, aimed at answering the question: how much can we infer from the voice about the face and vice versa? We study this task “in the wild”, employing the datasets that are now publicly available...
متن کامل