Unit selection in a concatenative speech synthesis system using a large speech database

By A. Hunt, Alan W. Black. 1996

Subfields: Natural Language Processing, Speech & Audio

How researchers see this paper

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

In Natural Language Processing

In Speech & Audio

Citations in the corpus

Jukebox: A Generative Model for Music (2020, background): “Many text-to-speech (TTS) systems rely on highly engineered features (Klatt, 1980), carefully curated sound segments (Hunt & Black, 1996), statistical parametric modeling (Zen et al., 2009), and often complex pipelines as described in (Arık et al., 2017).”
WaveNet: A Generative Model for Raw Audio (2016, methodology): “…speech synthesis part; non-parametric, example-based approach known as concatenative speech synthesis (Moulines & Charpentier, 1990; Sagisaka et al., 1992; Hunt & Black, 1996), and parametric, model-based approach known as statistical parametric speech synthesis (Yoshimura, 2002; Zen et al., 2009).”
Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions (2017, background): “Concatenative synthesis with unit selection, the process of stitching small units of pre-recorded waveforms together [2, 3] was the state-of-the-art for many years.”
Neural Speech Synthesis with Transformer Network (2018, methodology): “Concatenative TTS systems (Hunt and Black 1996; Black and Taylor 1997) and stitch them by some algorithms such as Viterbi (Viterbi 1967) followed by signal process methods (Charpentier and Stella 1986; Verhelst and Roelands 1993) to generate new waves.”
Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (2023, background): “Within the scope of text-to-speech (TTS), concatenative synthesis (Moulines & Charpentier, 1990; Hunt & Black, 1996) and statistical parametric synthesis (Yoshimura et al., 1999) were the prevailing approaches.”
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].”
Adversarial Audio Synthesis (2018, background): “In concatenative systems, audio is generated by sequencing small, prerecorded portions of speech from a phonetically-indexed dictionary (Moulines & Charpentier, 1990; Hunt & Black, 1996).”
DelightfulTTS: The Microsoft Speech Synthesis System for Blizzard Challenge 2021 (2021, methodology): “Rich unit selection based concatenative speech synthesis [8] and statistical parametric speech synthesis like HMM/DNN/LSTM based models [9, 10, 11], have been the most popular methods in the past.”
Listening while speaking: Speech chain by deep learning (2017, methodology): “In a similar direction, TTS technology development has gradually shifted from the foundation of a rule-based system using waveform coding and an analysis-synthesis method [7, 8, 9] to a waveform unit concatenation approach [10, 11] and a more flexible approach using the statistical modeling of hidde”
Speaker-Dependent WaveNet Vocoder (2017, methodology): “One of the representative techniques is concatenative synthesis [1, 2].”
Introduction to Digital Speech Processing (2007, unknown): “natural speech [31, 50, 108].”
Automatic Building of Synthetic Voices from Audio Books (2010, background): “The possibility of storing more than one example of a diphone unit, due to increase in storage and computation capabilities, has led to development of unit selection synthesis [Hunt and Black, 1996].”
Representation Mixing for TTS Synthesis (2018, unknown): “TTS synthesis [1] focuses on building systems which can generate speech (often in the form of an audio feature sequence a), given a set of linguistic sequences, l.”
Analysis and synthesis of intonation using the Tilt model. (2000, background): “Recently muc attention has been given to the notion of having large nu bers of voices in synthesis systems~Hunt and Black, 1996!.”
Recent Advances in Google Real-Time HMM-Driven Unit Selection Synthesizer (2016, background): “The goal is to find a sequence u∗ that minimizes the conventional unit selection cost [7] as described below.”
Speaker Adaptation in DNN-Based Speech Synthesis Using d-Vectors (2017, methodology): “Statistical parametric speech synthesis (SPSS) [1] has advantages when compared with speech synthesis methods based on unit selection and concatenation [2].”
Connectionist multivariate density-estimation and its application to speech synthesis (2016, background): “Unit-selection systems (Hunt and Black, 1996) store a database of natural speech and generate synthetic speech by pasting together segments of natural speech whose phonetic labels are most similar to those of the text to be synthesized.”
Generative adversarial network-based postfilter for statistical parametric speech synthesis (2016, background): “In the field of speech synthesis, statistical parametric speech synthesis based on hidden Markov models (HMMs) [1] or deep neural networks (DNNs) [2] has become popular since it provides various advantages over concatenative speech synthesis [3], such as the flexibility to control its voice characte”
Automatically clustering similar units for unit selection in speech synthesis (1997, methodology): “Expanding on [7], we use an acoustic vector which comprises Mel frequency cepstrum coefficients, F , power, and delta cepstrum, F and power.”
Machine Speech Chain (2020, methodology): “…technology has gradually shifted from the foundation of a rule-based system using waveform coding and an analysis-synthesis method [7]– [9] to waveform unit concatenation [10], [11] and a more ﬂexible approach using the statistical modeling of a hidden semi-Markov model-GMM (HSMM-GMM) [12], [13].”