Parallel WaveNet: Fast High-Fidelity Speech Synthesis
نویسندگان
چکیده
The recently-developed WaveNet architecture [27] is the current state of the art in realistic speech synthesis, consistently rated as more natural sounding for many different languages than any previous system. However, because WaveNet relies on sequential generation of one audio sample at a time, it is poorly suited to today’s massively parallel computers, and therefore hard to deploy in a real-time production setting. This paper introduces Probability Density Distillation, a new method for training a parallel feed-forward network from a trained WaveNet with no significant difference in quality. The resulting system is capable of generating high-fidelity speech samples at more than 20 times faster than real-time, and is deployed online by Google Assistant, including serving multiple English and Japanese voices.
منابع مشابه
WaveNet: A Generative Model for Raw Audio
This paper introduces WaveNet, a deep neural network for generating raw audio waveforms. The model is fully probabilistic and autoregressive, with the predictive distribution for each audio sample conditioned on all previous ones; nonetheless we show that it can be efficiently trained on data with tens of thousands of samples per second of audio. When applied to text-to-speech, it yields state-...
متن کاملSpeaker-Dependent WaveNet Vocoder
In this study, we propose a speaker-dependent WaveNet vocoder, a method of synthesizing speech waveforms with WaveNet, by utilizing acoustic features from existing vocoder as auxiliary features of WaveNet. It is expected that WaveNet can learn a sample-by-sample correspondence between speech waveform and acoustic features. The advantage of the proposed method is that it does not require (1) exp...
متن کاملNatural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinio...
متن کاملHybridnet: a Hybrid Neural Architecture to Speed-up Autoregressive Models
This paper introduces HybridNet, a hybrid neural network to speed-up autoregressive models for raw audio waveform generation. As an example, we propose a hybrid model that combines an autoregressive network named WaveNet and a conventional LSTM model to address speech synthesis. Instead of generating one sample per time-step, the proposed HybridNet generates multiple samples per time-step by ex...
متن کاملText-to-speech Synthesis System based on Wavenet
In this project, we focus on building a novel parametric TTS system. Our model is based on WaveNet(Oord et al, 2016), a deep neural network introduced by DeepMind in late 2016 for generating raw audio waveforms. It is fully probabilistic, with the predictive distribution for each audio sample conditioned on all previous samples. The model introduces the idea of convolutional layer into TTS task...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1711.10433 شماره
صفحات -
تاریخ انتشار 2017