An HMM-based speech synthesis system applied to English

By Keiichi Tokuda, Heiga Zen, Alan W. Black. Proceedings of 2002 IEEE Workshop on Speech Synthesis, 2002., 2002

406 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/63a4573652ddba4791c9a5ff02529e924c8dd65c

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

Citations in the corpus

Speech Synthesis Based on Hidden Markov Models (2013, methodology): “In addition to phoneme information, HMM-based speech synthesis uses various linguistic contexts such as lexical stress, pitch accent, tone, and part-of-speech information for the context-dependent modeling of HMMs [54].”
Parameter Generation Algorithms for Text-To-Speech Synthesis with Recurrent Neural Networks (2018, background): “AM based on Hidden Markov Models (HMMs) has been extensively studied for more than a decade [4, 5].”
Development of HMM-based Indonesian Speech Synthesis (2008, background): “It uses the same subject-verb-object word order used in English.”
The effects of part–of–speech tagging on text–to–speech synthesis for resource–scarce languages (2010, methodology): “The HTS context labels utilise by default a set of linguistic features defined in [87] and included in the demonstration package.”
Robust Speaker-Adaptive HMM-Based Text-to-Speech Synthesis (2009, neutral): “contain phonetic, segment-level, syllable-level, word-level and utterance-level features [49].”
Expressive TTS Training With Frame and Style Reconstruction Loss (2020, background): “The early studies of modeling speaking styles are carried out on Hidden Markov Models (HMM) [9], [66], where we can synthesize speech with an intermediate speaking style between two speakers through model interpolation [67].”
Statistical parametric speech synthesis using deep neural networks (2013, background): “In a typical system, there are normally around 50 different types of contexts [12].”
Deep mixture density networks for acoustic modeling in statistical parametric speech synthesis (2014, background): “In a typical implementation, there are normally around 50 different types of linguistic features [16], which is much more than those used in acoustic modeling for speech recognition.”
Unidirectional long short-term memory recurrent neural network with recurrent output layer for low-latency speech synthesis (2015, background): “A typical system uses around 50 different types of linguistic features [8].”
Statistical Parametric Speech Synthesis (2007, methodology): “…et al., 2006; Fares et al., 2008), Catalan (Bonafonte et al., 2008), Croatian (Martincic-Ipsic and Ipsic, 2006), Dzongkha (Sherpa et al., 2008), US (Tokuda et al., 2002b), UK, Scottish, Canadian, Indian, and South African English, Farsi (Homayoun-pour and Mehdi, 2004), Finnish (Ojala, 2006; Vainio…”
A Speech Parameter Generation Algorithm Considering Global Variance for HMM-Based Speech Synthesis (2007, methodology): “As one of the statistical synthesis meth-ods, we focus on the HMM-based speech synthesis method [5][6].”