Dataset · Hugging Face

liva-ai/swahili-asr

Name: swahili-asr
Creator: liva-ai
License: cc-by-4.0
Keywords: task_categories:automatic-speech-recognition, language:sw, license:cc-by-4.0, size_categories:n<1K, format:audiofolder, modality:audio, modality:text, library:datasets, library:mlcroissant, region:us, speech, conversational, diarization, code-switching, sheng

Swahili-ASR: 1400-Hour Conversational Swahili Speech Corpus Swahili-ASR is a 1400-hour conversational Swahili speech corpus with fully human-generated verbatim transcripts, produced by native Swahili speakers. The entire transcription process was built and managed in-house by the Liva AI team. Transcripts include millisecond-accurate speaker diarization with no overlapping timestamps between speakers, and capture spontaneous speech features such as disfluencies, interruptions… See the full description on the dataset page: https://huggingface.co/datasets/liva-ai/swahili-asr.

View on Hugging FaceReport an issue

Author / organization	liva-ai
License (as declared)	cc-by-4.0
Downloads	158 (as reported by Hugging Face)
Likes	0
Created	1 October 2026
Last updated at source	1 October 2026

speechconversationaldiarizationcode-switchingsheng

Embed this entry

<iframe src="https://silytics.si/embed/item/ef6dbfda-0027-48f8-a423-00aec731fb9d" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 3 days

158 now (+4 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from liva-ai

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.