paper-with-me

홈 › Papers

Rapid Speaker Adaptation in Low Resource Text to Speech Systems using Synthetic Data and Transfer learning

2023-12-02 · Raviraj Joshi, Nikesh Garera

Text-to-speech (TTS) systems are being built using end-to-end deep learning approaches. However, these systems require huge amounts of training data. We present our approach to built production quality TTS and perform speaker adaptation in extremely low resource settings. We propose a transfer learning approach using high-resource language data and synthetically generated data. We transfer the learnings from the out-domain high-resource English language. Further, we make use of out-of-the-box single-speaker TTS in the target language to generate in-domain synthetic data. We employ a three-step approach to train a high-quality single-speaker TTS system in a low-resource Indian language Hindi. We use a Tacotron2 like setup with a spectrogram prediction network and a waveglow vocoder. The Tacotron2 acoustic model is trained on English data, followed by synthetic Hindi data from the existing TTS system. Finally, the decoder of this model is fine-tuned on only 3 hours of target Hindi speaker data to enable rapid speaker adaptation. We show the importance of this dual pre-training and decoder-only fine-tuning using subjective MOS evaluation. Using transfer learning from high-resource language and synthetic corpus we present a low-cost solution to train a custom TTS model.

📄 PDF Abstract BibTeX arXiv:2312.01107

Code (0)

등록된 구현이 없습니다.

Tasks

Decodertext-to-speechText to SpeechTransfer Learning

Methods 이 논문이 사용한 방법론

Normalizing Flows Normalizing Flows are a method for constructing complex distributions by transforming a probability density through a series of invertible mappings. By repeatedly applying…
Affine Coupling 설명 없음
Invertible 1x1 Convolution The Invertible 1x1 Convolution is a type of convolution used in flow-based generative models that reverses the ordering of…
WaveGlow WaveGlow is a flow-based generative model that generates audio by sampling from a distribution. Specifically samples are taken from a zero mean spherical Gaussian with the…

Similar Papers 제목 키워드 기반

Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation

2021-06-06 · Dongchan Min, Dong Bok Lee, Eunho Yang, Sung Ju Hwang

With rapid progress in neural text-to-speech (TTS) models, personalized speech generation is now in high demand for many applications. For practical applicability, a TTS model should generate high-quality speech with onl…

text-to-speechText to Speech

On-the-Fly Feature Based Rapid Speaker Adaptation for Dysarthric and Elderly Speech Recognition

2022-03-28 · Mengzhe Geng, Xurong Xie, Rongfeng Su, Jianwei Yu 외

Accurate recognition of dysarthric and elderly speech remain challenging tasks to date. Speaker-level heterogeneity attributed to accent or gender, when aggregated with age and speech impairment, create large diversity a…

Diversityspeech-recognitionSpeech Recognition

VieSpeaker: A Large-Scale Vietnamese Speaker Recognition Dataset Beyond Visual Dependency

2026-06-23 · Viet Hoang Pham, Tran Trung Nguyen, Bao Thu Ho, Phuong Tuan Dat 외 arxiv

Speaker recognition has advanced rapidly with large-scale training datasets, yet Vietnamese remains under-resourced, with existing corpora limited in scale and acoustic diversity. Most large-scale datasets rely on facial…

Speaker Recognition

Multimodal speech synthesis architecture for unsupervised speaker adaptation

2018-08-20 · Hieu-Thi Luong, Junichi Yamagishi

This paper proposes a new architecture for speaker adaptation of multi-speaker neural-network speech synthesis systems, in which an unseen speaker's voice can be built using a relatively small amount of speech data witho…

Speech Synthesis

Residual Adapters for Few-Shot Text-to-Speech Speaker Adaptation

2022-10-28 · Nobuyuki Morioka, Heiga Zen, Nanxin Chen, Yu Zhang 외

Adapting a neural text-to-speech (TTS) model to a target speaker typically involves fine-tuning most if not all of the parameters of a pretrained multi-speaker backbone model. However, serving hundreds of fine-tuned neur…

text-to-speechText to Speech