Spectral voice conversion for text-to-speech synthesis

By A. Kain, Michael W. Macon. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Speech & Audio

Citations in the corpus

Learning Latent Representations for Speech Generation and Transformation (2017, background): “Understanding the influence of these factors on the speech signal is an important problem, which can be used for a wide variety of applications, including, but not limited to adaptation and data augmentation for speech recognition [1, 2], voice conversion [3, 4, 5], and speech compression [6].”
Many-to-Many Voice Transformer Network (2020, methodology): “ation contained in a given utterance such as the perceived identity of a speaker while preserving linguistic information. Potential applications of this technique include speaker-identity modiﬁcation [1], speaking aids [2], [3], speech enhancement [4]–[6], and accent conversion [7]. Many conventiona”
Multi-target Voice Conversion without Parallel Data by Adversarially Learning Disentangled Audio Representations (2018, background): “Examples of acoustic domains may include speaker identity [1, 2, 3, 4, 5, 6], speaking style, accent, emotion [7] or some other properties orthogonal to the linguistic content.”
Voice Conversion Using Deep Neural Networks With Layer-Wise Generative Training (2014, background): “The authors are with the National Engineering Laboratory of Speech and Language Information Processing, University of Science and Technology of China, Hefei 230027, China (e-mail: chenlh@ustc.edu.cn; zhling@ustc.edu.cn; ljliu037@mail.ustc.edu.cn; lrdai@ustc.edu.cn).”
An Overview of Voice Conversion and Its Challenges: From Statistical Modeling to Deep Learning (2020, methodology): “Therefore, subjective evaluations, such as MOS and similarity score, are also conducted.”
Spoofing and countermeasures for speaker verification: A survey (2015, methodology): “Gaussian mixture model (GMM) based approaches, which improve on the hard clustering of VQ methods were proposed in Kain and Macon (1998), Stylianou et al. (1998) and Toda et al., 2007 to implement a weighted linear conversion function.”
DNN-based stochastic postfilter for HMM-based speech synthesis (2014, methodology): “In voice conversion [7] this is typically done with a Gaussian mixture model (GMM) but a DNN was chosen here instead due to its abilities to model highly correlated and high dimensional data, allowing us to conduct spectral shaping directly in the spectral domain.”
Conditional restricted Boltzmann machine for voice conversion (2013, methodology): “inal proposed in [12], as our baseline method for comparison.”
NAUTILUS: A Versatile Voice Cloning System (2020, background): “In its broader definition, speech synthesis can refer to all kinds of speech generation interfaces like voice conversion (VC) [2], video-to-speech [3], [4], and others [5]–[7].”
Emotional Voice Conversion: Theory, Databases and ESD (2021, background): “Thus, spectrum conversion has been the focus [27, 28, 29].”
StarGAN-VC: non-parallel many-to-many Voice Conversion Using Star Generative Adversarial Networks (2018, methodology): “This technique can be applied to various tasks such as speaker-identity modification for text-to-speech (TTS) systems [1], speaking assistance [2, 3], speech enhancement [4–6], and pronunciation conversion [7].”
WaveNet Vocoder with Limited Training Data for Voice Conversion (2018, methodology): “In this approach, the acoustic features extracted from the source speech are ﬁrst mapped toward the target speaker using a conversion model, which could be a Gaussian mixture model (GMM) [1, 2, 3], an artiﬁcial neural network (ANN) [4, 5] and so on.”
Voice conversion in high-order eigen space using deep belief nets (2013, methodology): “The VC techniques have been applied to various tasks, such as speech enhancement [1], emotion conversion [2], speaking assistance [3], and other applications [4, 5].”
Voice Morphing that improves TTS quality using an optimal dynamic frequency warping-and-weighting transform (2016, methodology): “Standard statistical approaches have the form of GMM-based transformations [3, 4].”
CycleGAN-VC: Non-parallel Voice Conversion Using Cycle-Consistent Adversarial Networks (2018, methodology): “This technique can be applied to various tasks such as speaker-identity modification for text-to-speech systems [2], speaking assistance [3]–[5], emotion/expressiveness conversion [6], [7], and pronunciation conversion [8].”
A Deep Generative Architecture for Postfiltering in Statistical Parametric Speech Synthesis (2015, methodology): “In voice conversion [18] this is typically done with a Gaussian mixture model (GMM) but a DNN was chosen here instead due to its abilities to handle highly correlated and high-dimensional data, allowing us to conduct spectral shaping directly in the spectral amplitude domain.”