Review of text-to-speech conversion for English.

By D. Klatt. 1987

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

A Survey on Neural Speech Synthesis (2021, methodology): “The early computer-based speech synthesis methods include articulatory synthesis [52, 294], formant synthesis [293, 5, 167, 168], and concatenative synthesis [247, 235, 291, 124, 26].”
A Survey on Audio Diffusion Models: Text To Speech Synthesis and Enhancement in Generative AI (2023, unknown): “, formant synthesis [40, 41, 95] and concatenative synthesis [29, 69, 75]), statistical parametric speech synthesis (SPSS)-based methods [115, 116, 133], and neural network-based stage [122, 135, 136].”
Unsupervised Learning for Expressive Speech Synthesis (2017, methodology): “Some historical reference models for duration prediction are the Klatt model, by Klatt (1973), which is a deterministic model derived from corpus measuring where duration is calculated applying a set of factors determined by aspects like phone position, context, etc.”
Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis (2018, background): “…relating to speech articulation and pronunciation; the text-to-speech aspect was simply a set of rules for how these knobs were to be moved in response to phonemes extracted from text [15], and the resulting parameter trajectories could be manually edited in order to alter pronunciation.”
Speech synthesis techniques. A survey (2011, background): “However we do not yet know the optimal rule to do this [3].”
A review of verbal and non-verbal human-robot interactive communication (2014, background): “…were reputed to have been given a voice [3], and despite the fact that the ﬁrst general-purpose electronic speech synthesizer was developed by Noriko Omeda in Japan in 1968 [10], it wasnt until the early 1990’s that conversational robots such as MAIA [11], RHINO [12], and AESOP [13] appeared.”
Text-to-Speech Synthesis (2009, background): “Klatt’s 1987 article,Review of Text-to-Speech Conversion for English [255] is an excellent source for further reading on first generation systems.”
Hierarchical representation and estimation of prosody using continuous wavelet transform (2017, background): “The phonologically based ones stem from the so called Autosegmental Metrical theory [4] which is based on the three-dimensional phonology developed in [5, 6] as noted in [7].”
An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era (2022, methodology): “This paradigm was met with severe challenges in the modeling of articulatory behavior and was abandoned for a simpler, source-filter model that lends itself better to parameter control: formant synthesis [56], [57].”
Combining a vector space representation of linguistic context with a deep neural network for text-to-speech synthesis (2013, background): “These representations might be created in a number of ways, from the hand-coded rules of a formant synthesizer [1, 2] to the complex multi-layered structures typically found in systems such as Festival [3].”
A Review of Deep Learning Based Speech Synthesis (2019, methodology): “With the development of speech synthesis technologies, from the previous formant based parametric synthesis [1,2], waveform concatenation based methods [3–5] to the current statistical parametric speech synthesis (SPSS) [6], the intelligibility and naturalness of the synthesized speech have been…”
Synthesis of the Singing Voice by Performance Sampling and Spectral Models (2007, background): “…the one of the formant synthesizers (e.g. [17], CHANT system [18]), considered to be pseudo-physical models because even though they are mainly spectral models they make use of the source / filter decomposition which considers voice to be the result of a glottal excitation waveform (i.e. the…”
The derivation of prosody for text-to-speech from prosodic sentence structure☆ (1992, background): “Quite often, this drastic step results in obviously incorrect suprasegmental cues (Klatt, 1987).”
Introduction to Digital Speech Processing (2007, background): “A detailed survey of progress in text-to-speech conversion up to 1987 is given in the review paper by Klatt [65].”
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].”