The ICSI Meeting Corpus

By Adam L. Janin, D. Baron, Jane Edwards, D. Ellis, David Gelbart, Nelson Morgan, B. Peskin, T. Pfau, Elizabeth Shriberg, Andreas Stolcke and 1 more. 2003

923 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing, Evaluation & Benchmarks

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 Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

SCROLLS: Standardized CompaRison Over Long Language Sequences (2022, background): “The corpus covers academic group meetings at the International Computer Science Institute (Janin et al., 2003), 7 industrial product meetings for designing a remote control (Car-letta et al., 2005), and committee meetings of the Welsh 8 and Canadian 9 Parliaments, dealing with a variety of public…”
CHiME-6 Challenge: Tackling Multispeaker Speech Recognition for Unsegmented Recordings (2020, neutral): “Early attempts in this direction can be traced back to the ICSI [26], CHIL [27], and AMI [28] meeting corpora, the LLSEC [29] and COSINE [30] face-to-face interaction corpora, and the Shefﬁeld Wargames corpus [31].”
VoxCeleb: A Large-Scale Speaker Identification Dataset (2017, background): “ELSDSR [26] Clean Speech 22 198 MIT Mobile [21] Mobile Devices - 88 7,884 SWB [27] Telephony - 3,114 33,039 POLYCOST [17] Telephony - 133 1,285 ‡ ICSI Meeting Corpus [23] Meetings - 53 922 Forensic Comparison [22] Telephony 552 1,264 ANDOSL [18] Clean speech - 204 33,900 TIMIT [28] † Clean speech - ”
Argument Mining: A Survey (2020, methodology): “In the training stage, the system was used to process the first ten dialogues of the International Computer Science Institute meetings corpus (Janin et al. 2003) containing a total number of 98,523 words and 13,803 turns.”
End-to-End Neural Speaker Diarization with Permutation-Free Objectives (2019, background): “It also helps automatic speech recognition performance in multi-speaker conversation scenarios in meetings (ICSI [3,4], AMI [5,6]) and home environments (CHiME5 [6–10]).”
SLURP: A Spoken Language Understanding Resource Package (2020, background): “Multi-ASR is a large-scale system estimated from publicly available acoustic data pooled together – Acoustic data including, among others, LibriSpeech (Panayotov et al., 2015), Switchboard (Godfrey et al., 1992), Fisher (Cieri et al., 2004), CommonVoice (Ardila et al., 2019), AMI (Carletta, 2007) an”
Discovering hierarchical speech features using convolutional non-negative matrix factorization (2003, methodology): “To test the proposed feature extraction method, I used speech data from the ICSI meeting corpus [15].”
The Automatic Speech recogition In Reverberant Environments (ASpIRE) challenge (2015, methodology): “The International Computer Science Institute (ICSI) meeting room [3] study described in [4] used Efron’s bootstrap [5] to analyze the sources of error in an STT system; when the conditions were matched (even if far-field), it was observed that model errors dominated, but in mismatched conditions, fe”
The CHIL audiovisual corpus for lecture and meeting analysis inside smart rooms (2007, background): “Indeed, the vast majority of such publicly available corpora focuses on the audio modality alone, mainly aiming at speech technology development, such as the ICSI (Janin et al. 2003) and ISL (Burger et al. 2002) meeting data sets.”
SummScreen: A Dataset for Abstractive Screenplay Summarization (2021, background): “Colored boxes indicate utterances belonging to the same conversations. et al., 2015; Rush et al., 2015; Narayan et al., 2018; Grusky et al., 2018), online forums (Völske et al., 2017), meeting dialogues (Janin et al., 2003; Car-letta et al., 2005), and webpages (Chen et al., 2020).”
Long story short - Global unsupervised models for keyphrase based meeting summarization (2010, methodology): “For the experiments described in this work, we used manual and ASR transcripts of the ICSI (Janin et al., 2003) and AMI (McCowan et al., 2005) meeting corpora.”
The design and collection of COSINE, a multi-microphone in situ speech corpus recorded in noisy environments (2012, background): “The ICSI [15] and AMI [16] meeting corpora contain microphone array recordings of multi-party conversations in indoor environments.”
Discourse Segmentation of Multi-Party Conversation (2003, methodology): “We have evaluated our segmenter on the ICSI Meeting corpus (Janin et al., 2003).”
The ICSI Meeting Recorder Dialog Act (MRDA) Corpus (2004, methodology): “The ICSI Meeting Corpus data is described in detail in [9].”
Learning Bidirectional Intent Embeddings by Convolutional Deep Structured Semantic Models for Spoken Language Understanding (2015, methodology): “The dataset is from the ICSI meeting corpus [20], where 22 meetings previously used as test and dev sets are included for the actionable item detection task [21].”
Large-Scale Pre-Training of End-to-End Multi-Talker ASR for Meeting Transcription with Single Distant Microphone (2021, background): “Meeting transcription with a distant microphone has been widely studied as one of the most challenging problems for automatic speech recognition (ASR) [1, 2, 3].”
Sequential Short-Text Classification with Recurrent and Convolutional Neural Networks (2016, background): “• MRDA: ICSI Meeting Recorder Dialog Act Corpus (Janin et al., 2003; Shriberg et al., 2004).”
MediaSum: A Large-scale Media Interview Dataset for Dialogue Summarization (2021, methodology): “Speciﬁcally, we take the pre-trained model UniLM (Dong et al., 2019), ﬁne-tune it on M EDIA S UM , and then train it on datasets for meeting and dialogue summarization: AMI (McCowan et al., 2005), ICSI (Janin et al., 2003) and SAMSum (Gliwa et al., 2019).”
A French Corpus for Distant-Microphone Speech Processing in Real Homes (2016, neutral): “…and application scenarios such as voice command for cars [7–9] and in public spaces [10], automatic transcription of lectures [11], meetings [12–14], dialogues [15,16] and other public gatherings [17,18], and automatic transcription of noisy or overlapped speech in broadcast media [19,20].”