A Sliding-Window Approach to Automatic Creation of Meeting Minutes

Koay, J.; Roustai, Alexander; Dai, Xiaojin; Liu, Fei

doi:10.18653/v1/2021.naacl-srw.10

Cited by 20 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Almost all commercial STT services incorporate the ITN to increase the readability for users and automatic downstream processes. We choose four STT services that provide the audio transcription as the baseline systems including an inhouse Speech-to-Text service, Google Cloud Speech-to-Text service 3 , Microsoft Azure Speech-to-Text service 4 , and IBM Watson Speech-to-Text service 5 . Note we take these STT services as a black box, which means that we input the audio file and get the readable text from the service output in one shot.…”

Section: B Baseline Methodsmentioning

confidence: 99%

“…Automatic speech transcription that is highly readable for humans is required for applications like automatic subtitle generation [2,3] and meeting minutes generation [4,5], while machine translation [6,7], dialogue systems [8,9], voice search [10,11], voice question answering [12,13], and many other applications require highly readable transcriptions to generate the best machine response. If the system is unable to deal with deficiencies in speech transcription, it will have a substantial negative impact on the application users' experience.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Speech Recognition Post-Processing for Readability: Task, Dataset and a Two-Stage Pre-Trained Approach

Liao

2022

IEEE Access

View full text Add to dashboard Cite

Nowadays Automatic Speech Recognition (ASR) systems can accurately recognize which words are said. However, due to the disfluency, grammatical error, and other phenomena in spontaneous speech, the verbatim transcription of ASR impairs its readability, which is crucial for human comprehension and downstream tasks processing that need to understand the meaning and purpose of what is spoken. In this work, we formulate the ASR post-processing for readability (APR) as a sequence-to-sequence text generation problem that aims to transform the incorrect and noisy ASR output into readable text for humans and downstream tasks. We leverage the Metadata Extraction (MDE) corpus to construct a task-specific dataset for our study. To solve the problem of too little training data, we propose a novel data augmentation method that synthesizes large-scale training data from the grammatical error correction dataset. We propose a model based on the pre-trained language model to perform the APR task and train the model with a two-stage training strategy to better exploit the augmented data. On the constructed test set, our approach outperforms the best baseline system by a large margin of 17.53 on BLEU and 13.26 on readability-aware WER (RA-WER). The human evaluation also shows that our model can generate more human-readable transcripts than the baseline method.INDEX TERMS automatic post-editing, ASR post-processing for readability, data augmentation, pretrained language model, natural language processing

show abstract

Section: B Baseline Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Automatic Speech Recognition Post-Processing for Readability: Task, Dataset and a Two-Stage Pre-Trained Approach

Liao

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…While earlier works focus on extractive methods that create summaries by directly selecting and concatenating unedited sentences from source text, the development of neural networks has encouraged a growing trend in abstractive methods that implement encoder-decoder architectures on source text to gener-ate summaries (Shang et al 2018;Rennard et al 2023). The lengthy, multi-speaker, spoken-language natures of meeting text pose many challenges for summarization, several strategies are proposed to address different aspects of those challenges (Li et al 2019;Zhu et al 2020;Koay et al 2021;Zou et al 2021).…”

Section: Related Workmentioning

confidence: 99%

Learning from Failure: Improving Meeting Summarization without Good Samples

Wang,

Zhao,

Peng

2024

AAAI

View full text Add to dashboard Cite

Existing methods aligning language models with various human needs are reliant heavily on high-quality and task-specific data. However, industrial deployment of task-specific language models often encounter challenges in the availability of appropriate training samples. Taking meeting summarization for instance, public datasets are scarce, and private corpora are also hard to obtain due to privacy issues or resource-demanding annotation. To improve meeting summarization in the absence of positively-rated (i.e., ``good'') samples, we propose Score Tuning, a cold start tuning framework that leverages bad samples of distinguishable degrees to incrementally enhance the performance of summary generation without an initial presence of good samples. Our method utilizes asynchronous and numerical human feedback that measure the quality of generated summaries. Formulating data into triplets of (transcript, summary, score), our approach instructs a pre-trained model to learn the association between summary qualities and human-rated scores and hence to generate better summaries corresponding to higher scores. The experiment results show that our method is effective in improving meeting summarization on both English and Chinese corpora while requiring less annotated data and training resources compared to existing alignment methods. Additionally, we also preliminarily explore the transferability of our approach in machine translation tasks and demonstrate its potential for future development and usage in other domains.

show abstract

“…• Divide-and-Summarize: we prompt GPT-3.5 to obtain individual summary of each utterance, and then integrate summaries to form a complete summary (Koay et al, 2021).…”

Section: Macbertmentioning

confidence: 99%

ORCHID: A Chinese Debate Corpus for Target-Independent Stance Detection and Argumentative Dialogue Summarization

Zhao,

Wang,

Peng

2023

Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Dialogue agents have been receiving increasing attention for years, and this trend has been further boosted by the recent progress of large language models (LLMs). Stance detection and dialogue summarization are two core tasks of dialogue agents in application scenarios that involve argumentative dialogues. However, research on these tasks is limited by the insufficiency of public datasets, especially for non-English languages. To address this language resource gap in Chinese, we present OR-CHID (Oral Chinese Debate), the first Chinese dataset for benchmarking target-independent stance detection and debate summarization. Our dataset consists of 1,218 real-world debates that were conducted in Chinese on 476 unique topics, containing 2,436 stance-specific summaries and 14,133 fully annotated utterances. Besides providing a versatile testbed for future research, we also conduct an empirical study on the dataset and propose an integrated task. The results show the challenging nature of the dataset and suggest a potential of incorporating stance detection in summarization for argumentative dialogue. 1

show abstract

A Sliding-Window Approach to Automatic Creation of Meeting Minutes

Cited by 20 publications

References 33 publications

Automatic Speech Recognition Post-Processing for Readability: Task, Dataset and a Two-Stage Pre-Trained Approach

Automatic Speech Recognition Post-Processing for Readability: Task, Dataset and a Two-Stage Pre-Trained Approach

Learning from Failure: Improving Meeting Summarization without Good Samples

ORCHID: A Chinese Debate Corpus for Target-Independent Stance Detection and Argumentative Dialogue Summarization

Contact Info

Product

Resources

About