Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

SEQUENCE SEGMENTATION USING JOINT RNN AND STRUCTURED PREDICTION MODELS.

Yossi Adi¹, Joseph Keshet¹, Emily Cibelli²

¹Department of Computer Science, Bar-Ilan University, Ramat-Gan, Israel.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)

|October 17, 2017

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Unique Challenges of Multidisciplinary Clinical Science: Perspectives from a Multidisciplinary Team.

Clinical psychological science : a journal of the Association for Psychological Science·2026

Same author

Speech as an objective measure of psychomotor dysfunction in major depressive disorder: validation from non-speech motor measures.

Psychiatry research·2026

Same author

Open-vocabulary Keyword Spotting with Hyper-Matched Filters for Small Footprint Devices.

Computer speech & language·2026

Same author

Treatment of Haemophilia A Without Inhibitors: Real-World Treatment Patterns and Clinical Outcomes in the US.

Haemophilia : the official journal of the World Federation of Hemophilia·2026

Same author

Automatic Measurement of Voice Onset Time and Prevoicing using Recurrent Neural Networks.

Interspeech·2026

Same author

How does a deep neural network look at lexical stress in English words?

The Journal of the Acoustical Society of America·2026

Same journal

MAP Image Recovery with Guarantees using Locally Convex Multi-Scale Energy (LC-MUSE) Model.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2026

Same journal

EARLY DETECTION OF COGNITIVE DECLINE USING VOICE ASSISTANT COMMANDS.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2025

Same journal

CROSS-DOMAIN DIFFUSION BASED SPEECH ENHANCEMENT FOR VERY NOISY SPEECH.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2025

Same journal

CROSS-DOMAIN SPEECH ENHANCEMENT WITH A NEURAL CASCADE ARCHITECTURE.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2025

Same journal

ESTIMATING DIRECTED SPECTRAL INFORMATION FLOW BETWEEN MULTI-RESOLUTION TIME SERIES.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2025

Same journal

NEURAL CASCADE ARCHITECTURE FOR JOINT ACOUSTIC ECHO AND NOISE SUPPRESSION.

Proceedings of the ... IEEE International Conference on Acoustics, Speech, and Signal Processing. ICASSP (Conference)·2025

See all related articles

This study introduces a novel neural network for sequence segmentation in speech processing. The proposed model achieves state-of-the-art results in phonetic tasks like word and voice onset time segmentation.

Area of Science:

Speech Processing
Computational Linguistics
Machine Learning

Background:

Sequence segmentation is crucial for analyzing speech data.
Existing methods for speech segmentation have limitations.
Neural network approaches offer potential for improved segmentation accuracy.

Purpose of the Study:

To develop a simple and effective algorithm for sequence segmentation in speech processing.
To propose a novel neural architecture combining recurrent neural networks and structured prediction.
To evaluate the proposed method on phonetic segmentation tasks.

Main Methods:

A joint training approach for a recurrent neural network (RNN) module and a structured prediction model.
Utilizing RNN outputs as feature functions for the structured model.

Keywords:

Sequence segmentation recurrent neural networks (RNNs)structured prediction voice onset time word segmentation

Related Experiment Videos

Employing a task-specific structured loss function for training.

Main Results:

The proposed model demonstrated superior performance compared to previous methods.
State-of-the-art results were achieved on word segmentation tasks.
State-of-the-art results were achieved on voice onset time segmentation tasks.

Conclusions:

The proposed neural architecture is effective for sequence segmentation in speech processing.
The joint training of RNN and structured prediction models yields significant improvements.
This method offers a robust solution for phonetic segmentation challenges.