Dataset · Hugging Face

changelinglab/thchs30-segment

Name: thchs30-segment
Creator: changelinglab
License: apache-2.0
Keywords: task_categories:automatic-speech-recognition, language:zh, license:apache-2.0, size_categories:10K<n<100K, format:parquet, modality:audio, modality:text, library:datasets, library:dask, library:polars, library:mlcroissant, arxiv:1512.01882, region:us, speech, phone-alignment

THCHS-30 Segment Mandarin Chinese read-speech corpus with phone-level time alignments. Suitable for training and evaluating phone recognition and phonetic segmentation models. Sources Audio: THCHS-30 (OpenSLR 18) by Dong Wang, Xuewei Zhang, Zhiyong Zhang (Tsinghua University, 2015). Phone alignments: anyspeech/THCHS-30-alignments. Splits Split Utterances train 10,000 val 893 test 2,495 Splits follow the original OpenSLR 18 directory… See the full description on the dataset page: https://huggingface.co/datasets/changelinglab/thchs30-segment.

View on Hugging FaceReport an issue

Author / organization	changelinglab
License (as declared)	apache-2.0
Downloads	600 (as reported by Hugging Face)
Likes	0
Created	12 April 2026
Last updated at source	12 April 2026

speechphone-alignmentsegmentationmandarin

Embed this entry

<iframe src="https://silytics.si/embed/item/d14558e7-3634-4e9f-b942-f5f073169db6" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from changelinglab

changelinglab/cv-v1.0-segmentDataset
changelinglab/easycall-dysarthriaDataset
changelinglab/librispeech-segmentDataset
changelinglab/speechaccentarchive-prDataset
changelinglab/PBEBenchDataset
changelinglab/PBEBench-LiteDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.