Didispeech: A Large Scale Mandarin Speech Corpus

By Tingwei Guo, Cheng Wen, Dongwei Jiang, Ne Luo, Ruixiong Zhang, Shuaijiang Zhao, Wubo Li, Cheng Gong, Wei Zou, Kun Han and 1 more. 2020

63 citations recorded by the source corpus.

This paper introduces a new open-sourced Mandarin speech corpus, called DiDiSpeech. It consists of about 800 hours of speech data at 48kHz sampling rate from 6000 speakers and the corresponding texts. All speech data in the corpus is recorded in quiet environment and is suitable for various speech processing tasks, such as voice conversion, multi-speaker text-to-speech and automatic speech recognition. We conduct experiments with multiple speech tasks and evaluate the performance, showing that it is promising to use the corpus for both academic research and practical application. The corpus is available at https://outreach.didichuxing.com/research/opendata/.

Subfields: Artificial Intelligence

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. Xinyu Yang
2. Zhaoxi Mu
3. Yizhuo Dong
4. Lei Xie
5. Linquan Liu
6. Ziyang Ma
7. Yuping Wang
8. Zhikang Niu

Citations in the corpus

• A Survey on Neural Speech Synthesis (2021, neutral): “1 Mandarin DiDiSpeech-1 [97] 572 4500 48 Mandarin DiDiSpeech-2 [97] 227 1500 48 Mandarin JSUT [308] 10 1 48 Japanese KazakhTTS [237] 93 2 44.”
• Review of end-to-end speech synthesis technology based on deep learning (2021, background): “DiDiSpeech(8) [67] is a large open source Mandarin speech corpus released by DiDi Chuxing company.”
• Opencpop: A High-Quality Open Source Chinese Popular Song Corpus for Singing Voice Synthesis (2022, methodology): “The end-to-end TTS technology allows to train a model only with audio files and paired textual transcriptions, such as corpora LJ-Speech [25], VCTK [26], Aishell-3 [27], and DiDiSpeech [28].”
• Seed-TTS: A Family of High-Quality Versatile Speech Generation Models (2024, methodology): “Specifically, we employ 1,000 samples from the Common Voice dataset [Ardila et al., 2019] and 2,000 samples from the DiDiSpeech dataset [Guo et al., 2021].”
• Mega-TTS 2: Zero-Shot Text-to-Speech with Arbitrary Length Speech Prompts (2023, unknown): “…(Chen et al., 2021a) 8.1M 396.9M 9.6k LibriLight (Kahn et al., 2020) 10.0M 652.3M 17.9k LibriTTS (Zen et al., 2019) 375.1k 23.6M 573.2 WenetSpeech (Zhang et al., 2022) 14.0M 325.5M 9.3K AISHELL-3 (Shi et al., 2020) 88.0k 2.2M 85.0 Didi 500 (Guo et al., 2021) 52 prosodic prompt u rhy and u flat .”
• F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching (2024, methodology): “Three test sets are adopted for evaluation, which are LibriSpeech-PC test-clean [58], Seed-TTS test-en [39] with 1088 samples from Common Voice [59], and Seed-TTS test-zh with 2020 samples from DiDiSpeech [60] 3 .”