Statistical approach to vocal tract transfer function estimation based on factor analyzed trajectory HMM

By Tomoki Toda, Keiichi Tokuda. 2008

30 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

1. Junichi Yamagishi
2. Heiga Zen
3. Tomoki Toda
4. Keiichi Tokuda
5. Alan W. Black
6. Masanori Morise
7. Yoshihiko Nankaku
8. Zhenhua Ling

In Signal Processing

1. Junichi Yamagishi
2. Heiga Zen
3. Tomoki Toda
4. Keiichi Tokuda
5. Alan W. Black
6. Yoshihiko Nankaku
7. Masanori Morise
8. Zhenhua Ling

Citations in the corpus

• Integration of Spectral Feature Extraction and Modeling for HMM-Based Speech Synthesis (2014, methodology): “The vocal tract transfer function (VTTF) estimation of a speech signal based on a factor analyzed (FA) trajectory HMMs [19] is closely related to the proposed method in terms of the direct modeling of speech observation.”
• Error Evaluation of an F0-Adaptive Spectral Envelope Estimator in Robustness against the Additive Noise and F0 Error (2015, background): “The reliable spectral envelope is also important in statistical parametric speech synthesis [11], [12], and a spectral envelope estimator using a statistical approach [13] has been proposed to improve the sound quality.”
• Directly modeling speech waveforms by neural networks for statistical parametric speech synthesis (2015, methodology): “There are a couple of attempts to integrate feature extraction and acoustic model training into a single framework, e.g. the log spectral distortion-version of minimum generation error training (MGE­ LSD) [11], statistical vocoder (STAV OCO) [12], waveform-level statistical model [13], and mel-cepst”
• Statistical Parametric Speech Synthesis (2007, methodology): “Acero integrated formant analysis (Acero, 1999), Toda and Tokuda incorporated cepstral analysis (Toda and Tokuda, 2008), and Wu and Tokuda combined LSP parameter extraction (Wu and Tokuda, 2009).”
• A spectral envelope estimation method based on F0-adaptive multi-frame integration analysis (2012, methodology): “Several methods for statistically estimating the envelope from multiple frames have been studied in efforts to deal with this problem [28] [29] [30], and this multi-frame approach has also been used in a method estimating spectral envelopes of vocal from music sound mixtures [31].”
• CheapTrick, a spectral envelope estimator for high-quality speech synthesis (2015, methodology): “Alternatively, a spectral envelope estimation method using a statistical approach (Toda and Tokuda, 2008) was also proposed for the statistical parametric speech synthesis.”
• Speech Synthesis Based on Hidden Markov Models (2013, methodology): “proposed a statistical method for estimating the vocal tract transfer function from a speech signal based on the maximum a posteriori criterion [148].”
• Statistical parametric speech synthesis with joint estimation of acoustic and excitation model parameters (2010, methodology): “In the proposed approach spectral parameter extraction is integrated in the joint training of acoustic and excitation models, in a similar way to the spectral analysis based on factor analyzed trajectory HMM of [3].”
• Directly modeling voiced and unvoiced components in speech waveforms by neural networks (2016, background): “There have been a couple of attempts to integrate acoustic feature extraction into acoustic modeling, such as the log spectral distortion-version of minimum generation error training [7], statistical vocoder [8], waveform-level statistical model [9], and melcepstral analysis-integrated hidden Markov”