You need to agree to share your contact 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.

Synthetic Tajik speech corpus (OmniVoice)

Single-speaker synthetic Tajik speech, generated 2026-09-22..24.

corpus clips hours
books 324,861 397.92
khazina 116,254 657.90
law 192,530 634.71
total 633,645 1690.53

Sources: khazina (book readings), law (legal texts), books (book texts).

Licence and intended use

CC-BY-NC-4.0. Research use only. Generated with an OmniVoice LoRA (omnivoice-lora-sl-panther2); OmniVoice is CC-BY-NC, so this corpus inherits the non-commercial restriction. Every manifest row carries license_note: "CC-BY-NC-derived, research-only, never Sado product". This data must not be used in a commercial product.

Layout

<corpus>/manifests/manifest_shard*.jsonl   text labels, one JSON per clip
<corpus>/audio/*.tar                       24 kHz mono wavs, sharded

Manifest fields: clip_id, text, wav, duration, ref_voice, generator, synthetic, rms_dbfs, normalization_target_dbfs, peak_ceiling_dbfs, normalization_method, license_note, source_row.

All audio is loudness-normalised to -16 dBFS RMS with a -1 dBFS peak ceiling. It is synthetic: a single cloned reference voice (preset_female_pro), not real speakers.

Downloads last month
70