The fifth 'CHiME' Speech Separation and Recognition Challenge: Dataset, task and baselines

Abstract : The CHiME challenge series aims to advance robust automatic speech recognition (ASR) technology by promoting research at the interface of speech and language processing, signal processing , and machine learning. This paper introduces the 5th CHiME Challenge, which considers the task of distant multi-microphone conversational ASR in real home environments. Speech material was elicited using a dinner party scenario with efforts taken to capture data that is representative of natural conversational speech and recorded by 6 Kinect microphone arrays and 4 binaural microphone pairs. The challenge features a single-array track and a multiple-array track and, for each track, distinct rankings will be produced for systems focusing on robustness with respect to distant-microphone capture vs. systems attempting to address all aspects of the task including conversational language modeling. We discuss the rationale for the challenge and provide a detailed description of the data collection procedure, the task, and the baseline systems for array synchronization, speech enhancement, and conventional and end-to-end ASR.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [37 references]  Display  Hide  Download

https://hal.inria.fr/hal-01744021
Contributor : Emmanuel Vincent <>
Submitted on : Tuesday, March 27, 2018 - 2:24:21 AM
Last modification on : Wednesday, April 3, 2019 - 1:22:57 AM
Document(s) archivé(s) le : Thursday, September 13, 2018 - 9:35:14 AM

Files

barker_IS18.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01744021, version 1
  • ARXIV : 1803.10609

Collections

Citation

Jon Barker, Shinji Watanabe, Emmanuel Vincent, Jan Trmal. The fifth 'CHiME' Speech Separation and Recognition Challenge: Dataset, task and baselines. Interspeech 2018 - 19th Annual Conference of the International Speech Communication Association, Sep 2018, Hyderabad, India. ⟨hal-01744021⟩

Share

Metrics

Record views

208

Files downloads

497