Downloads · 30 days
28
17% of all-time downloads
saudi-arabic-tts
aszfezf/saudi-arabic-tts
saudi-arabic-tts is a text-to-speech model from aszfezf. Use it when you need text read aloud. It is set up for chatterbox. The card lists the license as mit.
Fine-tuned Chatterbox TTS T3 component on Arabic TTS data.
All-time downloads
162
Public
Parameters
507M
4.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.2 GB · 74%
At a glance
- Task
- Text-to-Speech
- Library
- chatterbox
- License
- mit
- Model type
- qwen2
- Access
- Public
- Created
- Feb 24, 2026
- Updated
- Feb 25, 2026
- SHA
- f35dc411
Base models
Model card
From the Hugging Face model README
Arabic Fine-tuned Chatterbox TTS
Fine-tuned Chatterbox TTS T3 component on Arabic TTS data.
Model Details
- Base Model: ResembleAI/chatterbox (multilingual)
- Fine-tuned Component: T3 (text-to-speech token transformer, 535M params)
- Training Data: NeoBoy/arabic-tts-wav-24k (Arabic diacritized TTS)
- Language: Arabic (ar)
Usage
from chatterbox import ChatterboxMultilingualTTS
from safetensors.torch import load_file as load_safetensors
from huggingface_hub import snapshot_download
from pathlib import Path
ckpt_dir = Path(snapshot_download("aszfezf/saudi-arabic-tts"))
model = ChatterboxMultilingualTTS.from_local(ckpt_dir, device="cuda")
wav = model.generate("مرحبا بالعالم", language_id="ar")
- Task
- Text-to-Speech
- Library
- chatterbox
- Type
- qwen2
- License
- mit
- Languages
- ar
- Created
- Feb 24, 2026
- Updated
- Feb 25, 2026