Search results for: Gustav Eje Henter

Items from 1 to 13 out of 13 results

article

Style‐Controllable Speech‐Driven Gesture Synthesis Using Normalising Flows

Simon Alexanderson, Gustav Eje Henter, Taras Kucherenko, Jonas Beskow

Computer Graphics Forum > 39 > 2 > 487 - 496

Automatic synthesis of realistic gestures promises to transform the fields of animation, avatars and communicative agents. In off‐line applications, novel tools can alter the role of an animator to that of a director, who provides only high‐level input for the desired animation; a learned network then translates these instructions into an appropriate sequence of body poses. In interactive scenarios,...

article

Investigating different representations for modeling and controlling multiple emotions in DNN-based speech synthesis

Jaime Lorenzo-Trueba, Gustav Eje Henter, Shinji Takaki, Junichi Yamagishi, more

Speech Communication > 2018 > 99 > C > 135-143

In this paper, we investigate the simultaneous modeling of multiple emotions in DNN-based expressive speech synthesis, and how to represent the emotional labels, such as emotional class and strength, for this task. Our goal is to answer two questions: First, what is the best way to annotate speech data with multiple emotions – should we use the labels that the speaker intended to express, or labels...

chapter

Adapting and controlling DNN-based speech synthesis using input codes

Hieu-Thi Luong, Shinji Takaki, Gustav Eje Henter, Junichi Yamagishi

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) > 4905 - 4909

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Methods for adapting and controlling the characteristics of output speech are important topics in speech synthesis. In this work, we investigated the performance of DNN-based text-to-speech systems that in parallel to conventional text input also take speaker, gender, and age codes as inputs, in order to 1) perform multi-speaker synthesis, 2) perform speaker adaptation using small amounts of target-speaker...

chapter

Median-based generation of synthetic speech durations using a non-parametric approach

Srikanth Ronanki, Oliver Watts, Simon King, Gustav Eje Henter

2016 IEEE Spoken Language Technology Workshop (SLT) > 686 - 692

2016 IEEE Spoken Language Technology Workshop (SLT)

This paper proposes a new approach to duration modelling for statistical parametric speech synthesis in which a recurrent statistical model is trained to output a phone transition probability at each timestep (acoustic frame). Unlike conventional approaches to duration modelling - which assume that duration distributions have a particular form (e.g., a Gaussian) and use the mean of that distribution...

article

Minimum Entropy Rate Simplification of Stochastic Processes

Gustav Eje Henter, W. Bastiaan Kleijn

IEEE Transactions on Pattern Analysis and Machine Intelligence > 2016 > 38 > 12 > 2487 - 2500

We propose minimum entropy rate simplification (MERS), an information-theoretic, parameterization-independent framework for simplifying generative models of stochastic processes. Applications include improving model quality for sampling tasks by concentrating the probability mass on the most characteristic and accurately described behaviors while de-emphasizing the tails, and obtaining clean models...

article

Bayesian Analysis of Phoneme Confusion Matrices

Leijon Leijon, Gustav Eje Henter, Martin Dahlquist

IEEE/ACM Transactions on Audio, Speech, and Language Processing > 2016 > 24 > 3 > 469 - 482

This paper presents a parametric Bayesian approach to the statistical analysis of phoneme confusion matrices measured for groups of individual listeners in one or more test conditions. Two different bias problems in conventional estimation of mutual information are analyzed and explained theoretically. Evaluations with synthetic datasets indicate that the proposed Bayesian method can give satisfactory...

chapter

From HMMS to DNNS: Where do the improvements come from?

Oliver Watts, Gustav Eje Henter, Thomas Merritt, Zhizheng Wu, more

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) > 5505 - 5509

ICASSP 2016 - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Deep neural networks (DNNs) have recently been the focus of much text-to-speech research as a replacement for decision trees and hidden Markov models (HMMs) in statistical parametric synthesis systems. Performance improvements have been reported; however, the configuration of systems evaluated makes it impossible to judge how much of the improvement is due to the new machine learning methods, and...

chapter

Testing the consistency assumption: Pronunciation variant forced alignment in read and spontaneous speech synthesis

Rasmus Dali, Sandrine Brognaux, Korin Richmond, Cassia Valentini-Botinhao, more

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) > 5155 - 5159

ICASSP 2016 - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Forced alignment for speech synthesis traditionally aligns a phoneme sequence predetermined by the front-end text processing system. This sequence is not altered during alignment, i.e., it is forced, despite possibly being faulty. The consistency assumption is the assumption that these mistakes do not degrade models, as long as the mistakes are consistent across training and synthesis. We present...

chapter

Robust TTS duration modelling using DNNS

Gustav Eje Henter, Srikanth Ronanki, Oliver Watts, Mirjam Wester, more

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) > 5130 - 5134

ICASSP 2016 - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Accurate modelling and prediction of speech-sound durations is an important component in generating more natural synthetic speech. Deep neural networks (DNNs) offer a powerful modelling paradigm, and large, found corpora of natural and expressive speech are easy to acquire for training them. Unfortunately, found datasets are seldom subject to the quality-control that traditional synthesis methods...

article

Picking up the pieces: Causal states in noisy data, and how to recover them

Gustav Eje Henter, W. Bastiaan Kleijn

Pattern Recognition Letters > 2013 > 34 > 5 > 587-594

Automatic structure discovery is desirable in many Markov model applications where a good topology (states and transitions) is not known a priori. CSSR is an established pattern discovery algorithm for stationary and ergodic stochastic symbol sequences that learns a predictively optimal Markov representation consisting of so-called causal states. By means of a novel algebraic criterion, we prove that...

article

Maximizing Phoneme Recognition Accuracy for Enhanced Speech Intelligibility in Noise

Petko N. Petkov, Gustav Eje Henter, W. Bastiaan Kleijn

IEEE Transactions on Audio, Speech, and Language Processing > 2013 > 21 > 5 > 1035 - 1045

An effective measure of speech intelligibility is the probability of correct recognition of the transmitted message. We propose a speech pre-enhancement method based on matching the recognized text to the text of the original message. The selected criterion is accurately approximated by the probability of the correct transcription given an estimate of the noisy speech features. In the presence of...

chapter

Gaussian process dynamical models for nonparametric speech representation and synthesis

Gustav Eje Henter, Marcus R. Frean, W. Bastiaan Kleijn

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) > 4505 - 4508

ICASSP 2012 - 2012 IEEE International Conference on Acoustics, Speech and Signal Processing

We propose Gaussian process dynamical models (GPDMs) as a new, nonparametric paradigm in acoustic models of speech. These use multidimensional, continuous state-spaces to overcome familiar issues with discrete-state, HMM-based speech models. The added dimensions allow the state to represent and describe more than just temporal structure as systematic differences in mean, rather than as mere correlations...

article

Simplified probability models for generative tasks: A rate-distortion approach

Gustav Eje Henter, W. Bastiaan Kleijn

02010 00018th European Signal Processing Conference > 2010 > 1159 - 1163

2010 18th European Signal Processing Conference

We consider using sparse simplifications to denoise probabilistic sequence models for generative tasks such as speech synthesis. Our proposal is to find the least random model that remains close to the original one according to a KL-divergence constraint, a technique we call minimum entropy rate simplification (MERS). This produces a representation-independent framework for trading off simplicity...

Filter options

Publication date

Set your own date range

INFONA - science communication portal

Search results for: Gustav Eje Henter

Style‐Controllable Speech‐Driven Gesture Synthesis Using Normalising Flows

Investigating different representations for modeling and controlling multiple emotions in DNN-based speech synthesis

Adapting and controlling DNN-based speech synthesis using input codes

Median-based generation of synthetic speech durations using a non-parametric approach

Minimum Entropy Rate Simplification of Stochastic Processes

Bayesian Analysis of Phoneme Confusion Matrices

From HMMS to DNNS: Where do the improvements come from?

Testing the consistency assumption: Pronunciation variant forced alignment in read and spontaneous speech synthesis

Robust TTS duration modelling using DNNS

Picking up the pieces: Causal states in noisy data, and how to recover them

Maximizing Phoneme Recognition Accuracy for Enhanced Speech Intelligibility in Noise

Gaussian process dynamical models for nonparametric speech representation and synthesis

Simplified probability models for generative tasks: A rate-distortion approach

Filter options

Publication date

Publication type

Keywords

Data set

Journal

INFONA - science communication portal

Search results for: Gustav Eje Henter

Add recipient

Sending message cancelled

Are you sure you want to cancel sending this message?

Send message

Filter options

Publication date

Date range setting

Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.

Publication type

Keywords

Data set

Journal

Reporting an error / abuse

Sending the report failed

Accessibility options