COVAREP—A collaborative voice analysis repository for speech technologies
Speech processing algorithms are often developed demonstrating improvements over the
state-of-the-art, but sometimes at the cost of high complexity. This makes algorithm …
state-of-the-art, but sometimes at the cost of high complexity. This makes algorithm …
[HTML][HTML] A uniform phase representation for the harmonic model in speech synthesis applications
G Degottex, D Erro - EURASIP Journal on Audio, Speech, and Music …, 2014 - Springer
Feature-based vocoders, eg, STRAIGHT, offer a way to manipulate the perceived
characteristics of the speech signal in speech transformation and synthesis. For the …
characteristics of the speech signal in speech transformation and synthesis. For the …
Reversible speaker de-identification using pre-trained transformation functions
C Magarinos, P Lopez-Otero… - Computer Speech & …, 2017 - Elsevier
Speaker de-identification approaches must accomplish three main goals: universality,
naturalness and reversibility. The main drawback of the traditional approach to speaker de …
naturalness and reversibility. The main drawback of the traditional approach to speaker de …
A log domain pulse model for parametric speech synthesis
G Degottex, P Lanchantin… - IEEE/ACM Transactions on …, 2017 - ieeexplore.ieee.org
Most of the degradation in current Statistical Parametric Speech Synthesis (SPSS) results
from the form of the vocoder. One of the main causes of degradation is the reconstruction of …
from the form of the vocoder. One of the main causes of degradation is the reconstruction of …
Harmonics-to-noise ratio estimation with deterministically time-varying harmonic model for pathological voice signals
T Ikuma, B Story, AJ McWhorter, L Adkins… - The Journal of the …, 2022 - pubs.aip.org
The harmonics-to-noise ratio (HNR) and other spectral noise parameters are important in
clinical objective voice assessment as they could indicate the presence of nonharmonic …
clinical objective voice assessment as they could indicate the presence of nonharmonic …
[HTML][HTML] Neural speech-rate conversion with multispeaker WaveNet vocoder
Speech-rate conversion technology, which can expand or compress speech waveforms
while preserving the pitch of the sound, is traditionally realized by signal-processing-based …
while preserving the pitch of the sound, is traditionally realized by signal-processing-based …
[PDF][PDF] Fusion of multiple parameterisations for DNN-based sinusoidal speech synthesis with multi-task learning.
It has recently been shown that deep neural networks (DNN) can improve the quality of
statistical parametric speech synthesis (SPSS) when using a source-filter vocoder. Our own …
statistical parametric speech synthesis (SPSS) when using a source-filter vocoder. Our own …
Demodulated sound quality improvement for harmonic sounds in over-boosted parametric array loudspeaker
Y Geng, M Nakayama, T Nishiura - Applied Acoustics, 2022 - Elsevier
The parametric array loudspeaker (PAL) can realize sharp directivity utilizing the
straightness of ultrasound, but it suffers from low sound quality due to the frequency …
straightness of ultrasound, but it suffers from low sound quality due to the frequency …
Subjective and objective assessment of full bandwidth speech quality
JG Beerends, NMP Neumann… - … on Audio, Speech …, 2019 - ieeexplore.ieee.org
With the introduction of fullband speech coding the question arises what role frequency
components above 14 kHz play in speech quality assessment. On the one hand, our results …
components above 14 kHz play in speech quality assessment. On the one hand, our results …
Towards objective voice assessment: the diplophonia diagram
P Aichinger, I Roesner, B Schneider-Stickler… - Journal of voice, 2017 - Elsevier
Objectives Diplophonia is an often misinterpreted symptom of disordered voice, and needs
objectification. An audio signal processing algorithm for the detection of diplophonia is …
objectification. An audio signal processing algorithm for the detection of diplophonia is …