ComfyUI-SparkTTS: Advanced Text-to-Speech for ComfyUI

https://github.com/1038lab/ComfyUI-SparkTTS

Introducing our new member of our Comfyui TTS collections ComfyUI-SparkTTS: Advanced Text-to-Speech for ComfyUI

ComfyUI-SparkTTS is a custom node implementation of SparkTTS, an advanced text-to-speech system powered by large language models (LLMs). Designed for both research and production, it delivers highly accurate and natural-sounding speech synthesis.

If this node is helpful to you or if you like Our work, please give a ⭐ on github, It will be the greatest encouragement for our efforts!

Key Features:

Efficient & Streamlined – Built on Qwen2.5, removing the need for additional acoustic models.

Zero-Shot Voice Cloning – Replicates voices without training data, perfect for multilingual and code-switching scenarios.

Bilingual Support – Synthesizes speech in both Chinese and English with high fluency.

Customizable Speech – Adjust parameters like gender, pitch, and speaking rate for unique virtual speakers.

An additional key point is that this node has minimal dependencies, making it easy to install with virtually no setup requirements. It is compatible with both CPU and GPU, allowing for versatile operation.