Datasets:

disco-eth
/

WorldSpeech

Tasks:

Automatic Speech Recognition

Audio Classification

Modalities:

Languages:

Size:

ArXiv:

Tags:

DOI:

License:

Dataset card Files Files and versions

AAdonis commited on 20 days ago

Commit

e433d1e

·

verified ·

1 Parent(s): a8795eb

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -674,7 +674,7 @@ configs:
 # WorldSpeech
-**Highly multilingual, large-scale paired speech corpus assembled from public archives such as national parliaments, public broadcasters, and public-domain audiobooks across 107 language-country configs, with strong non-European and low-resource coverage.**
 Every segment contains a verified *human* transcript, an ASR transcript, the CER between them, a WADA-SNR estimate, and four DNSMOS-P.835 quality scores.

 # WorldSpeech
+**Highly multilingual, large-scale speech corpus with ground-truth aligned transcripts assembled from public archives such as national parliaments, public broadcasters, and public-domain audiobooks across 107 language-country configs, with strong non-European and low-resource coverage.**
 Every segment contains a verified *human* transcript, an ASR transcript, the CER between them, a WADA-SNR estimate, and four DNSMOS-P.835 quality scores.