Loquacious Set: 25,000 Hours of Transcribed and Diverse English Speech Recognition Data for Research and Commercial Use

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parcollet, Titouan, Tseng, Yuan, Zhang, Shucong, van Dalen, Rogier
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908381893099520
author Parcollet, Titouan
Tseng, Yuan
Zhang, Shucong
van Dalen, Rogier
author_facet Parcollet, Titouan
Tseng, Yuan
Zhang, Shucong
van Dalen, Rogier
contents Automatic speech recognition (ASR) research is driven by the availability of common datasets between industrial researchers and academics, encouraging comparisons and evaluations. LibriSpeech, despite its long success as an ASR benchmark, is now limited by its size and focus on clean, read speech, leading to near-zero word error rates. More recent datasets, including MOSEL, YODAS, Gigaspeech, OWSM, Libriheavy or People's Speech suffer from major limitations including licenses that researchers in the industry cannot use, unreliable transcriptions, incorrect audio data, or the lack of evaluation sets. This work presents the Loquacious Set, a 25,000-hour curated collection of commercially usable English speech. Featuring hundreds of thousands of speakers with diverse accents and a wide range of speech types (read, spontaneous, talks, clean, noisy), the Loquacious Set is designed to work for academics and researchers in the industry to build ASR systems in real-world scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21578
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Loquacious Set: 25,000 Hours of Transcribed and Diverse English Speech Recognition Data for Research and Commercial Use
Parcollet, Titouan
Tseng, Yuan
Zhang, Shucong
van Dalen, Rogier
Computation and Language
Sound
Audio and Speech Processing
Automatic speech recognition (ASR) research is driven by the availability of common datasets between industrial researchers and academics, encouraging comparisons and evaluations. LibriSpeech, despite its long success as an ASR benchmark, is now limited by its size and focus on clean, read speech, leading to near-zero word error rates. More recent datasets, including MOSEL, YODAS, Gigaspeech, OWSM, Libriheavy or People's Speech suffer from major limitations including licenses that researchers in the industry cannot use, unreliable transcriptions, incorrect audio data, or the lack of evaluation sets. This work presents the Loquacious Set, a 25,000-hour curated collection of commercially usable English speech. Featuring hundreds of thousands of speakers with diverse accents and a wide range of speech types (read, spontaneous, talks, clean, noisy), the Loquacious Set is designed to work for academics and researchers in the industry to build ASR systems in real-world scenarios.
title Loquacious Set: 25,000 Hours of Transcribed and Diverse English Speech Recognition Data for Research and Commercial Use
topic Computation and Language
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2505.21578