LibriSpeech

LibriSpeech is a corpus of approximately 1000 hours of 16kHz read English speech, prepared by Vassil Panayotov with the assistance of Daniel Povey. The data is derived from read audiobooks from the LibriVox project, and has been carefully segmented and aligned.
Acoustic models, trained on this data set, are available at [icefall](https://github.com/k2-fsa/icefall/tree/master/egs/librispeech) and language models, suitable for evaluation can be found at [openslr](http://www.openslr.org/11/).

For more information, see the paper "LibriSpeech: an ASR corpus based on public domain audio books", Vassil Panayotov, Guoguo Chen, Daniel Povey and Sanjeev Khudanpur, ICASSP 2015 [pdf](https://www.danielpovey.com/files/2015_icassp_librispeech.pdf)

LibriSpeech是一个包含约1000小时16kHz英语朗读语音的语料库，由瓦西里·帕纳约托夫（Vassil Panayotov）主导构建，丹尼尔·波维（Daniel Povey）提供协助。该数据集源自LibriVox项目的朗读有声读物，并经过严格的分段与对齐处理。
基于该数据集训练的声学模型（acoustic models）可从[icefall](https://github.com/k2-fsa/icefall/tree/master/egs/librispeech)获取，适用于评估任务的语言模型（language models）则可在[openslr](http://www.openslr.org/11/)下载。
如需获取更多相关信息，请参阅论文《LibriSpeech：基于公有领域有声读物的自动语音识别（Automatic Speech Recognition）语料库》（"LibriSpeech: an ASR corpus based on public domain audio books"），作者为瓦西里·帕纳约托夫、陈果果（Guoguo Chen）、丹尼尔·波维与桑吉夫·库丹普尔（Sanjeev Khudanpur），该论文发表于2015年国际声学、语音与信号处理会议（ICASSP 2015），[pdf](https://www.danielpovey.com/files/2015_icassp_librispeech.pdf)

搜集汇总

背景与挑战

背景概述

LibriSpeech是一个包含约1000小时16kHz英语朗读语音的语料库，数据来源于LibriVox项目的有声读物，并经过精确分段和对齐处理。该数据集还提供了配套的声学模型和语言模型，适用于自动语音识别（ASR）研究。