You need to agree to share your Hugging Face information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Piper-plus multilingual 7-language v8 — aggregate audit

This metadata-only audit describes ayousanz/piper-plus-multilingual-7lang-v8-dataset at revision d6774fdd1ea77f21fdd7105f63ce8973c01642ad. It contains no audio, transcripts, speaker identifiers, cache tensors, embeddings, or per-utterance paths.

The fixed revision has 342,855 training rows across 3,682 speakers plus a speaker-safe 1,200-row holdout with ten additional speakers. The seven observed language IDs are 0–5 and 7; ID 6 is unused although the config declares eight language slots. All nine split archive hashes match Hub LFS, and the essential archive contains exactly one 192-dimensional float32 embedding per training row.

The referenced source-rights manifest was due for review on 2026-08-18, omits three current source families, and has no entry whose stated selected counts exactly match v8. The provenance table records these gaps. No umbrella license is declared; each source's license and restrictions must be applied independently.

Downloads last month
50