Comprehensive Evaluation of Statistical Speech Waveform Synthesis

Merritt, Thomas; Putrycz, Bartosz; Nadolski, Adam; Ye, Tianjun; Korzekwa, Daniel; Dolecki, Wiktor; Drugman, Thomas; Klimkov, Viacheslav; Moinet, Alexis; Breen, Andrew P.; Kuklinski, Rafal; Ström, Nikko; Barra-Chicote, Roberto

doi:10.1109/slt.2018.8639556

Cited by 9 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On every test the positions of the systems on the panel were randomised. All the panels included the natural recordings as an upper anchor but similar to [20] subjects were not forced to assign the top score to any of the systems. All tests were conducted in Amazon Mechanical Turk.…”

Section: Subjective Evaluationmentioning

confidence: 99%

Effect of Data Reduction on Sequence-to-sequence Neural TTS

Latorre

Lachowicz

Lorenzo-Trueba

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Recent speech synthesis systems based on sampling from autoregressive neural networks models can generate speech almost undistinguishable from human recordings. However, these models require large amounts of data. This paper shows that the lack of data from one speaker can be compensated with data from other speakers. The naturalness of Tacotron2-like models trained on a blend of 5k utterances from 7 speakers is better than that of speaker dependent models trained on 15k utterances, but in terms of stability multi-speaker models are always more stable. We also demonstrate that models mixing only 1250 utterances from a target speaker with 5k utterances from another 6 speakers can produce significantly better quality than state-ofthe-art DNN-guided unit selection systems trained on more than 10 times the data from the target speaker.

show abstract

Section: Subjective Evaluationmentioning

confidence: 99%

Effect of Data Reduction on Sequence-to-sequence Neural TTS

Latorre

Lachowicz

Lorenzo-Trueba

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, many medical and healthcare devices based on deep learning technology have been proposed for tasks such as the detection of pathological voice [71], healthcare monitoring [72], heart disease prediction [73], detection and reconstruction of dysarthric speech [74], and speech waveform synthesis [75]. Through the application of deep learning with big data, we gain many benefits for healthcare applications compared with traditional approaches.…”

Section: The Existing Application Of Deep Learning Technology In Heal...mentioning

confidence: 99%

A Speech Command Control-Based Recognition System for Dysarthric Patients Based on Deep Learning Technology

et al. 2021

View full text Add to dashboard Cite

Voice control is an important way of controlling mobile devices; however, using it remains a challenge for dysarthric patients. Currently, there are many approaches, such as automatic speech recognition (ASR) systems, being used to help dysarthric patients control mobile devices. However, the large computation power requirement for the ASR system increases implementation costs. To alleviate this problem, this study proposed a convolution neural network (CNN) with a phonetic posteriorgram (PPG) speech feature system to recognize speech commands, called CNN–PPG; meanwhile, the CNN model with Mel-frequency cepstral coefficient (CNN–MFCC model) and ASR-based systems were used for comparison. The experiment results show that the CNN–PPG system provided 93.49% accuracy, better than the CNN–MFCC (65.67%) and ASR-based systems (89.59%). Additionally, the CNN–PPG used a smaller model size comprising only 54% parameter numbers compared with the ASR-based system; hence, the proposed system could reduce implementation costs for users. These findings suggest that the CNN–PPG system could augment a communication device to help dysarthric patients control the mobile device via speech commands in the future.

show abstract

“…We conducted MUSHRA perceptual test [37]. Every listener was presented with 6 versions of a given word at the same time, 5 reconstructions and one version of recorded speech.…”

Section: Reconstruction Of Dysarthric Speechmentioning

confidence: 99%

Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech

et al. 2019

Self Cite

View full text Add to dashboard Cite

We present a novel deep learning model for the detection and reconstruction of dysarthric speech. We train the model with a multi-task learning technique to jointly solve dysarthria detection and speech reconstruction tasks. The model key feature is a low-dimensional latent space that is meant to encode the properties of dysarthric speech. It is commonly believed that neural networks are black boxes that solve problems but do not provide interpretable outputs. On the contrary, we show that this latent space successfully encodes interpretable characteristics of dysarthria, is effective at detecting dysarthria, and that manipulation of the latent space allows the model to reconstruct healthy speech from dysarthric speech. This work can help patients and speech pathologists to improve their understanding of the condition, lead to more accurate diagnoses and aid in reconstructing healthy speech for afflicted patients.

show abstract

Comprehensive Evaluation of Statistical Speech Waveform Synthesis

Cited by 9 publications

References 25 publications

Effect of Data Reduction on Sequence-to-sequence Neural TTS

Effect of Data Reduction on Sequence-to-sequence Neural TTS

A Speech Command Control-Based Recognition System for Dysarthric Patients Based on Deep Learning Technology

Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech

Contact Info

Product

Resources

About