STRAIGHT, exploitation of the other aspect of VOCODER: Perceptually isomorphic decomposition of speech sounds

By Hideki Kawahara. 2006

326 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

In Signal Processing

Citations in the corpus

A Survey on Neural Speech Synthesis (2021, methodology): “In statistical parametric speech synthesis [409, 350, 408, 418, 351], LSP (line spectral pairs) [132], MCC (mel-cepstral coefficients) [79], MGC (mel-generalized coefficients) [349], F0 and BAP (band aperiodicities) [153, 154] are used as acoustic features, which can be easily converted into wavefor”
Voice Morphing that improves TTS quality using an optimal dynamic frequency warping-and-weighting transform (2016, methodology): “Source/target speaker spectra are extracted using analysis similar to STRAIGHT [25], excluding unvoiced frames as in [4, 16, 26].”
Technical foundations of TANDEM-STRAIGHT, a speech analysis, modification and synthesis framework (2011, methodology): “STRAIGHT, a speech analysis, modification, and synthesis framework (Kawahara et al 1999a), was originally designed to promote speech perception research by providing a tool to manipulate naturally sounding speech material in terms of perceptually relevant and precisely controllable physical paramete”
A Template-Based Approach for Speech Synthesis Intonation Generation Using LSTMs (2016, unknown): “Finally, waveforms were synthesised using the STRAIGHT vocoder [26] and normalised according to ITU P.”
Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis (2018, background): “Similarly, the waveform generator is traditionally a ﬁxed, designed component, for example STRAIGHT [74] or WORLD [75], to whose control interface the acoustic feature representation is tied.”
HiFiSinger: Towards High-Fidelity Neural Singing Voice Synthesis (2020, background): “Usually, the window size should cover 2∼8 times of the period of the fundamental frequency (Juvela et al., 2016; Kawahara, 2006).”
Vocaine the vocoder and applications in speech synthesis (2015, background): “The two most notable contributions here are STRAIGHT [14, 15] and AhoCoder [16] which is an STC/HNM variant.”
Emotional voice conversion using deep neural networks with MCC and F0 features (2016, methodology): “Although many researchers have adopted the F0 features for emotional VC [18][19], the F0 features used in these approaches were mostly extracted by the STRAIGHT [20].”
An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era (2022, background): “Some notable vocoders are WORLD [62] and STRAIGHT [63].”
An investigation of multi-speaker training for wavenet vocoder (2017, methodology): “These acoustic features are calculated using STRAIGHT analysis [20].”
Improve GAN-based Neural Vocoder using Pointwise Relativistic LeastSquare GAN (2021, background): “Compared with conventional vocoders([2], [3]), neural vocoders can significantly enhance the speech synthesis quality of the current text-to-speech (TTS) system.”
Adapting and controlling DNN-based speech synthesis using input codes (2017, unknown): “STRAIGHT [26] analysis was used to obtain 259-dimensional acoustic feature vectors every 5 ms, each comprising 60 mel-cepstral coefﬁcients (with a bilinear frequency warping parameter of 0.77), a linearly interpolated fundamental frequency in the mel scale, and 25-dimensional band aperiodicities,…”
Towards Robust Neural Vocoding for Speech Generation: A Survey (2019, background): “Traditionally, waveform can be vocoded from these acoustic or linguistic features using heuristic methods [9] or handcrafted vocoders [10, 11, 12].”
Improve GAN-based Neural Vocoder using Truncated Pointwise Relativistic Least Square GAN (2022, background): “Compared with conventional vocoders([18], [9]), neural vocoders can significantly enhance the speech synthesis quality of the current text-to-speech (TTS) and voice conversion (VC) systems.”
Tandem-STRAIGHT: A temporally stable power spectral representation for periodic signals and applications to interference-free spectrum, F0, and aperiodicity estimation (2008, methodology): “STRAIGHT [1], a speech analysis, modification, and synthesis system (as well as speech morphing based on it) is widely used in the speech research community [2, 3, 4, 5, 6].”
Excitation source analysis for high-quality speech manipulation systems based on an interference-free representation of group delay with minimum phase response compensation (2014, methodology): “STRAIGHT [1] and TANDEM-STRAIGHT [2] were designed to fulﬁll this requirement to provide ﬂexible tools for exploratory research of speech [3].”