AISHELL-4: An Open Source Dataset for Speech Enhancement, Separation, Recognition and Speaker Diarization in Conference Scenario

By Yihui Fu, Luyao Cheng, Shubo Lv, Yukai Jv, Yuxiang Kong, Zhuo Chen, Yanxin Hu, Lei Xie, Jian Wu, Hui Bu and 3 more. 2021

162 citations recorded by the source corpus.

In this paper, we present AISHELL-4, a sizable real-recorded Mandarin speech dataset collected by 8-channel circular microphone array for speech processing in conference scenario. The dataset consists of 211 recorded meeting sessions, each containing 4 to 8 speakers, with a total length of 120 hours. This dataset aims to bridge the advanced research on multi-speaker processing and the practical application scenario in three aspects. With real recorded meetings, AISHELL-4 provides realistic acoustics and rich natural speech characteristics in conversation such as short pause, speech overlap, quick speaker turn, noise, etc. Meanwhile, accurate transcription and speaker voice activity are provided for each meeting in AISHELL-4. This allows the researchers to explore different aspects in meeting processing, ranging from individual tasks such as speech front-end processing, speech recognition and speaker diarization, to multi-modality modeling and joint optimization of relevant tasks. Given most open source dataset for multi-speaker tasks are in English, AISHELL-4 is the only Mandarin dataset for conversation speech, providing additional value for data diversity in speech community. We also release a PyTorch-based training and evaluation framework as baseline system to promote reproducible research in this field.

Subfields: Artificial Intelligence, 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

1. Alexis Plaquet
2. Hervé Bredin

In Signal Processing

1. Alexis Plaquet
2. Hervé Bredin
3. Yuxin Peng

Citations in the corpus

• Make-A-Voice: Revisiting Voice Large Language Models as Scalable Multilingual and Multitask Learners (2024, neutral): “• Librilight (Kahn et al., 2020) contains 60K hours of unlabelled speech from audiobooks in English, and LibriSpeech (Panayotov et al., 2015), Lib-riTTS (Zen et al., 2019), Gigaspeech (Chen et al., 2021), AISHELL (Fu et al., 2021), VCTK (Veaux et al., 2017) datasets include transcriptions.”
• Pyannote.audio 2.1 Speaker Diarization Pipeline: Principle, Benchmark, and Recipe (2023, background): “• A growing number of freely available datasets such as AISHELL-4 [11], Albayzin/RTVE [10], AliMeeting [12], AMI [13], VoxConverse [9], Ego4D [14], or This American Life [15].”
• Powerset multi-class cross entropy loss for neural speaker diarization (2023, methodology): “We report performance on the official test sets of 9 different datasets: AISHELL-4 [12], AliMeeting [13], AMI [14] (with two variants), DIHARD III [21], Ego4D 2 [15], MSDWild [16], REPERE [17], and VoxConverse [22].”