ModelPrésentéExterne

Vekol Tts Ckb Edge

Piper (VITS) 22.05 kHz ONNX edge model for Central Kurdish (Sorani). Runs on CPU / edge devices.

Sourcehuggingface.co/RevgeAI/vekol-tts-ckb-edge/tree/64037ebd9dedeacb7b946752d474a3a265b3f853

TypeOtherVersionRevision 64037ebCClasse

Source du modèle

Type: Other
Taille du fichier: 73,5 MB
Version: vRevision 64037eb
Mis à jour: 10 oct. 2026
Téléchargements: 0
Source: Hugging Face

Description de la source

Vekol Piper edge (ckb)

Piper (VITS) 22.05 kHz ONNX edge model for Central Kurdish (Sorani). Runs on CPU / edge devices.

By Revge / Vekol ( ڕێڤگە / ڤەکۆڵ ).

Training data

Resumed from a Central Kurdish base, then trained on the Vekol voice plus teacher-distilled audio, with weak-letter balanced oversampling.

Stage	Data	Hours	Clips
Base	DB1 + DB2 (Central Kurdish corpora)	27.33 h	11,694
Fine-tune — real	Vekol voice	~59 min	397 sentences
Fine-tune — distilled	teacher (Vekol)	~3 h	~1,500

Weak letters (گ ڤ ح غ) are oversampled ×2; long-text tail-hallucination fixed in inference.

Usage

model.onnx + model.onnx.json with the Piper runtime, or the included vekol_tts.py helper.

Sources

1 source

Hugging FacePrincipalVersion 64037ebd9dedeacb7b946752d474a3a265b3f853

huggingface.co/RevgeAI/vekol-tts-ckb-edge/tree/64037ebd9dedeacb7b946752d474a3a265b3f853

Vérifié 22 août

Artefacts du modèle

1 artefact

model.onnx

onnx · 73,5 MB · SHA-256 92a30317e6e9…3f39 · Hugging Face

Télécharger

Extraits de sources

2 extraits

Readme22 aoûthuggingface.co/RevgeAI/vekol-tts-ckb-edge

--- license: cc-by-nc-4.0 language: [ckb] pipeline_tag: text-to-speech library_name: piper pretty_name: Vekol Piper edge (ckb) tags: [kurdish, sorani, central-kurdish, ckb, tts, text-to-speech, speech, piper, vits, onnx, edge] --- # Vekol Piper edge (ckb) Piper (VITS) 22.05 kHz ONNX edge model for Central Kurdish (Sorani). Runs on CPU / edge devices. By Revge / Vekol ( ڕێڤگە / ڤەکۆڵ ). ## Training data Resumed from a Central Kurdish base, then trained on the Vekol voice plus teacher-distilled audio, with weak-letter balanced oversampling. | Stage | Data | Hours | Clips | |---|---|---|---| | Base | DB1 + DB2 (Central Kurdish corpora) | 27.33 h | 11,694 | | Fine-tune — real | Vekol voice | ~59 min | 397 sentences | | Fine-tune — distilled | teacher (Vekol) | ~3 h | ~1,500 | Weak letters (گ ڤ ح غ) are oversampled ×2; long-text tail-hallucination fixed in inference. ## Usage `model.onnx` + `model.onnx.json` with the Piper runtime, or the included `vekol_tts.py` helper.

Note22 aoûthuggingface.co/RevgeAI/vekol-tts-ckb-edge

Source context: 0 downloads · 0 likes · Pipeline text-to-speech · Library piper · Repo RevgeAI/vekol-tts-ckb-edge