The HMM-based speech synthesis system (HTS) version 2.0

By Heiga Zen, Takashi Nose, Junichi Yamagishi, Shinji Sako, Takashi Masuko, Alan W. Black, Keiichi Tokuda. Speech Synthesis Workshop, 2007

635 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/091ba178d649ac0d7da617c5bdb37aba3a3385b3

How researchers see this paper

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

Citations in the corpus

Review of end-to-end speech synthesis technology based on deep learning (2021, methodology): “of TTS techniques from the formant-based methods [92, 97, 105, 106, 179, 224] to the unit selection-based waveform cascade methods [6, 24, 36, 59, 80, 143, 144], and to the hidden Markov model (HMM)-based statistical parametric speech synthesis (SPSS) methods [26, 94, 149, 178, 210, 241, 249, 250].”
FoundationTTS: Text-to-Speech for ASR Customization with Generative Language Model (2023, background): “Human speech naturally has a random sampling effect like white noise in source-filter vocoders [24, 25, 26, 27, 28] that status linguistic or acoustic condition is not enough to simulate.”
Implementing Prosodic Phrasing in Chinese End-to-end Speech Synthesis (2019, methodology): “At first we tried to leverage the full context label of the phone, as was used in the HMM-based TTS system [3].”
Phonological vocoding using artificial neural networks (2015, methodology): “The three-state, cross-word triphone models were trained with the HTS variant [18] of the HTK toolkit on the set of 56,231 utterances.”
Adapting and controlling DNN-based speech synthesis using input codes (2017, unknown): “A new DNN baseline sys-tem [11] added to the Blizzard Challenge 2016 1 also turned out to be signiﬁcantly better than the standard HMM-based baseline (that uses a toolkit called HTS [12]), again conﬁrming the speech quality improvements brought on by deep learning approaches.”
Unsupervised Learning for Expressive Speech Synthesis (2017, background): “For instance, Trilla and Alias (2013) already used sentiment analysis on sentence level for an expressive TTS. Vanmassenhove et al. (2016) also used sentiment combined with emotion labels for an HMM-based system.”
Composition of Deep and Spiking Neural Networks for Very Low Bit Rate Speech Coding (2016, methodology): “For HMM analysis models, we trained three-state, cross-word triphone models with the HTS variant [52] of the HTK toolkit on the WSJ training set.”
Prosody Transfer in Neural Text to Speech Using Global Pitch and Loudness Features (2019, background): “These frameworks have traditionally leveraged hidden Markov models (HMM) for parametric speech synthesis [6].”
A Neural Parametric Singing Synthesizer (2017, background): “3) [38] was followed, except for a somewhat simplified context dependency (just the two previous and two following phonemes).”
Speech Synthesis Based on Hidden Markov Models (2013, background): “Although many research groups have contributed to the recent progress in HMM-based speech synthesis, note that the description given here is somewhat biased toward implementation of the HMM-based Speech Synthesis System (HTS) [11], [40].”
GlotNet—A Raw Waveform Model for the Glottal Excitation in Statistical Parametric Speech Synthesis (2019, methodology): “The linguistic and acoustic features are aligned with the HMM-based speech synthesis system (HTS) [38] for training the neural net acoustic model.”
Polyphone Disambiguation for Mandarin Chinese Using Conditional Neural Network with Multi-level Embedding Features (2019, methodology): “The grapheme-to-phoneme (G2P) conversion is a fundamental front-end procedure in the Chinese Text-to-Speech (TTS) synthesis system, either the traditional HMM-based speech synthesis system [1, 2] or the End-to-End speech synthesis system [3, 4, 5, 6].”
Introducing AmuS: The Amused Speech Database (2017, methodology): “The implementation was done using the HTS toolkit [45].”