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

By Jon Barker, Shinji Watanabe, Emmanuel Vincent, Jan "Yenda" Trmal. 2018

729 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Evaluation & Benchmarks, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Evaluation & Benchmarks

In Speech & Audio

In Signal Processing

Citations in the corpus

CHiME-6 Challenge: Tackling Multispeaker Speech Recognition for Unsegmented Recordings (2020, methodology): “The CHiME-6 challenge, which builds upon CHiME-5 [34], targets the problem of distant microphone conversational speech recognition in everyday home environments.”
ESPnet: End-to-End Speech Processing Toolkit (2018, methodology): “With these state-of-theart end-to-end ASR techniques, ESPnet also provides a number of recipes for major ASR benchmarks including Wall Street Journal (WSJ) [18], Librispeech [19], TED-LIUM [20], Corpus of Spontaneous Japanese (CSJ) [21], AMI [22], HKUST Mandarin CTS [23], VoxForge [24], CHiME-4/5 [2”
A Comparative Study on Transformer vs RNN in Speech Applications (2019, background): “AISHELL [28] zh 170 read dev / test AURORA4 [29] (*) en 15 noisy read (dev 0330) A / B / C / D CSJ [30] ja 581 spontaneous eval1 / eval2 / eval3 CHiME4 [31] (*) en 108 noisy far-field multi-ch read dt05 simu / dt05 real / et05 simu / et05 real CHiME5 [32] en 40 noisy far-field multi-ch conversationa”
The STC System for the CHiME-6 Challenge (2020, methodology): “The details on the challenge are described in [2].”
Acoustic Modeling for Overlapping Speech Recognition: Jhu Chime-5 Challenge System (2019, methodology): “This section describes the CHiME-5 data and the baseline ASR system released by the challenge organizers [6].”
Front-end processing for the CHiME-5 dinner party scenario (2018, background): “The CHiME-5 challenge [10], introduces a database consisting only of real multi-channel recordings in a dinner party scenario.”
Unsupervised adaptation with domain separation networks for robust speech recognition (2017, methodology): “The CHiME-3 dataset is released with the 3rd CHiME speech Separation and Recognition Challenge [29], which incorporates the Wall Street Journal corpus sentences spoken in challenging noisy environments, recorded using a 6-channel tablet based microphone array.”
EasyCom: An Augmented Reality Dataset to Support Algorithms for Easy Communication in Noisy Environments (2021, methodology): “The CHiME-5 [8] and CHiME-6 [9] challenge datasets, released in 2018 and re-run in 2019, respectively, contain 32 audio channels of 4 participants having a dinner party in private homes.”
Neural Speaker Diarization with Speaker-Wise Chain Rule (2020, background): “Recent studies on the home-party scenario (CHiME-5 [10]) reported that speaker diarization helped improve automatic speech recognition performance [11–13].”
Survey on Deep Neural Networks in Speech and Vision Systems (2019, background): “CHiME 5 [150] is a speech recognition dataset that contains challenging speech recognition conditions including multiple speaker natural conversations.”
Hearing aid Research Data Set for Acoustic Environment Recognition (2020, methodology): “Interfering Speakers consists of audio from CHiME 5 [8] (the multi-microphone conversational speech dinner party).”