Neural TTS Voice Conversion

Kons, Zvi; Shechtman, Slava; Sorin, Alex; Hoory, Ron; Rabinovitz, Carmel; Morais, Edmilson

doi:10.1109/slt.2018.8639550

Cited by 9 publications

(16 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper we show that we can get a considerable quality improvement by modifying a TTS system that produced the WORLD vocoder parameters [6] to predict parameters for LPCNet [8]. As in the previous work [6], we conduct multiple adaptation experiments, applied on multiple VCTK voices [9] and show that the new system has much better quality and similarity to the target voices but can still run much faster than real-time in a single-CPU mode.…”

Section: Introductionmentioning

confidence: 87%

“…In the current work, the prosody generation and adaptation network follows the one presented in our previous work [6], where one can refer to for more details. It generates a 4dimensional prosody vector per TTS unit, comprising the unit's log-duration, initial log-pitch, final log-pitch and logenergy.…”

Section: Prosody Generatormentioning

confidence: 99%

“…In this architecture the prosody adaptation to unseen speaker is based on a Variational Auto Encoder (VAE) utterance prosody embedding, averaged over all the speaker utterances [6], as presented on figure 2. In the current work we used multi-speaker baseline models for prosody adaptation to unseen voices, as it resulted in better quality than the single speaker models.…”

Section: Prosody Generatormentioning

confidence: 99%

“…For each of those voices we built the following single speaker TTS systems: 1. A WORLD based system at 22KHz as described in [6].…”

Section: High Quality Voicesmentioning

confidence: 99%

“…In these tests a subject is presented with a pair of samples that convey different text messages. The subject is asked to rate their voice similarity, using a 4-point scale adopted from the Voice Conversion Challenge (VCC) [15] [ 16], and utilized in our previous experiments [6]. We performed two tests: one with only male voices and the second with only female voices.…”

Section: Voice Adaptationmentioning

confidence: 99%

See 4 more Smart Citations

High Quality, Lightweight and Adaptable TTS Using LPCNet

et al. 2019

Self Cite

View full text Add to dashboard Cite

We present a lightweight adaptable neural TTS system with high quality output. The system is composed of three separate neural network blocks: prosody prediction, acoustic feature prediction and Linear Prediction Coding Net as a neural vocoder. This system can synthesize speech with close to natural quality while running 3 times faster than real-time on a standard CPU.The modular setup of the system allows for simple adaptation to new voices with a small amount of data.We first demonstrate the ability of the system to produce high quality speech when trained on large, high quality datasets. Following that, we demonstrate its adaptability by mimicking unseen voices using 5 to 20 minutes long datasets with lower recording quality. Large scale Mean Opinion Score quality and similarity tests are presented, showing that the system can adapt to unseen voices with quality gap of 0.12 and similarity gap of 3% compared to natural speech for male voices and quality gap of 0.35 and similarity of gap of 9 % for female voices.

show abstract

Section: Introductionmentioning

confidence: 87%

Section: Prosody Generatormentioning

confidence: 99%

Section: Prosody Generatormentioning

confidence: 99%

“…For each of those voices we built the following single speaker TTS systems: 1. A WORLD based system at 22KHz as described in [6].…”

Section: High Quality Voicesmentioning

confidence: 99%

Section: Voice Adaptationmentioning

confidence: 99%

See 3 more Smart Citations

High Quality, Lightweight and Adaptable TTS Using LPCNet

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

Enhancing Speech Recognition for Parkinson’s Disease Patient Using Transfer Learning Technique

2021

J. Shanghai Jiaotong Univ. (Sci.)

View full text Add to dashboard Cite

Voice Conversion Based Augmentation and a Hybrid CNN-LSTM Model for Improving Speaker-Independent Keyword Recognition on Limited Datasets

Wubet

Lian

2022

IEEE Access

View full text Add to dashboard Cite

Keyword recognition is the basis of speech recognition, and its application is rapidly increasing in keyword spotting, robotics, and smart home surveillance. Because of these advanced applications, improving the accuracy of keyword recognition is crucial. In this paper, we proposed voice conversion (VC) -based augmentation to increase the limited training dataset and a fusion of a convolutional neural network (CNN) and long-short term memory (LSTM) model for robust speaker-independent isolated keyword recognition. Collecting and preparing a sufficient amount of voice data for speaker-independent speech recognition is a tedious and bulky task. To overcome this, we generated new raw voices from the original voices using an auxiliary classifier conditional variational autoencoder (ACVAE) method. In this study, the main intention of voice conversion is to obtain numerous and various human-like keywords' voices that are not identical to the source and target speakers' pronunciation. Parallel VC was used to accurately maintain the linguistic content. We examined the performance of the proposed voice conversion augmentation techniques using robust deep neural network algorithms. Original training data, excluding generated voice using other data augmentation and regularization techniques, were considered as the baseline. The results showed that incorporating voice conversion augmentation into the baseline augmentation techniques and applying the CNN-LSTM model improved the accuracy of isolated keyword recognition.

show abstract

Neural TTS Voice Conversion

Cited by 9 publications

References 4 publications

High Quality, Lightweight and Adaptable TTS Using LPCNet

High Quality, Lightweight and Adaptable TTS Using LPCNet

Enhancing Speech Recognition for Parkinson’s Disease Patient Using Transfer Learning Technique

Voice Conversion Based Augmentation and a Hybrid CNN-LSTM Model for Improving Speaker-Independent Keyword Recognition on Limited Datasets

Contact Info

Product

Resources

About