Text-to-Speech
GGUF
Persian
Farsi
Persian

In the Name of God

Chatterbox TTS Persian gguf

GGUF / ggml conversion of Thomcles/Chatterbox-TTS-Persian-Farsi for use with CrispStrobe/CrispASR.

Persian T3 (text-to-speech-tokens) variant of Chatterbox TTS.

demo audios:

"سلام! به آزمایش تبدیل متن به گفتار خوش آمدید."

"سه سیب سرخ روی سینی سیمی است"

"دیروز در تهران باران شد، امروز آفتابی است"

"قیمت لپ‌تاپ جدید من پنجاه میلیون تومان است."

Chatterbox is a full TTS pipeline: character tokenizer → T3 (30-layer Llama AR, 520M) → speech tokens → S3Gen (Conformer encoder + UNet1D CFM denoiser, 10 Euler steps) → HiFTGenerator vocoder (conv chains + Snake activations + iSTFT) → 24 kHz WAV.

How to use:

Two GGUF files are needed: the T3 model (text → speech tokens) and the S3Gen model (speech tokens → audio).

  1. Download CrispASR:

github.com/CrispStrobe/CrispASR/releases

  1. Download model files:

chatterbox-t3-fa-f16.gguf

chatterbox-s3gen-q8_0.gguf

  1. Synthesise with the built-in default voice:

./crispasr.exe --backend chatterbox -m chatterbox-t3-fa-f16.gguf --codec-model chatterbox-s3gen-q8_0.gguf --tts "این سیستم طراحی شده است تا به کاربر دسترسی سریع و آسان به اطلاعات مورد نیاز را بدهد." --tts-output out.wav

Downloads last month
399
GGUF
Model size
0.3B params
Architecture
chatterbox-s3gen
Hardware compatibility
Log In to add your hardware

8-bit

16-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for mazrba/Chatterbox-TTS-Persian-gguf

Quantized
(24)
this model

Datasets used to train mazrba/Chatterbox-TTS-Persian-gguf