Search Results for author:
Found 10 papers, 3 papers with code
Learning the joint distribution of two sequences using little or no paired data
We present a noisy channel generative model of two sequences, for example text and speech, which enables uncovering the association between the two modalities when limited paired data is available.
Global Normalization for Streaming Speech Recognition in a Modular Framework
We introduce the Globally Normalized Autoregressive Transducer (GNAT) for addressing the label bias problem in streaming speech recognition.
Speaker Generation
We call this task "speaker generation", and present TacoSpawn, a system that performs competitively at this task.
Non-saturating GAN training as divergence minimization
Non-saturating generative adversarial network (GAN) training is widely used and has continued to obtain groundbreaking results.
Properties of f-divergences and f-GAN training
In this technical report we describe some properties of f-divergences and f-GAN training.
Location-Relative Attention Mechanisms For Robust Long-Form Speech Synthesis
Despite the ability to produce human-level speech for in-domain text, attention-based end-to-end text-to-speech (TTS) systems suffer from text alignment failures that increase in frequency for out-of-domain text.
Semi-Supervised Generative Modeling for Controllable Speech Synthesis
We present a novel generative model that combines state-of-the-art neural text-to-speech (TTS) with semi-supervised probabilistic latent variable models.
The divergences minimized by non-saturating GAN training
The original variant is theoretically easier to study, but for GANs the alternative variant performs better in practice.
Effective Use of Variational Embedding Capacity in Expressive End-to-End Speech Synthesis
Recent work has explored sequence-to-sequence latent variable models for expressive speech synthesis (supporting control and transfer of prosody and style), but has not presented a coherent framework for understanding the trade-offs between the competing methods.
Optimizing expected word error rate via sampling for speech recognition
State-level minimum Bayes risk (sMBR) training has become the de facto standard for sequence-level training of speech recognition acoustic models.
