ComfyUI-KittenTTS and ComfyUI-SparkTTS
These are competing custom ComfyUI nodes, each providing a distinct text-to-speech model (KittenTTS vs. SparkTTS) for generating audio from text within the ComfyUI workflow.
About ComfyUI-KittenTTS
Saganaki22/ComfyUI-KittenTTS
😻 A simple ComfyUI custom node for KittenTTS - an ultra-lightweight text-to-speech model. Works on CUDA and CPU.
Wraps KittenTTS models (19–80MB) as a single ComfyUI node with automatic model caching and dynamic ONNX Runtime selection for CPU/GPU inference. Features phoneme-based synthesis via espeak-ng, 8 voice options, adjustable speech speed, and optional stereo output, with models ranging from nano-int8 quantization to 80M parameter variants for quality-performance tradeoffs.
About ComfyUI-SparkTTS
1038lab/ComfyUI-SparkTTS
ComfyUI-SparkTTS is a custom ComfyUI node implementation of SparkTTS, an advanced text-to-speech system that harnesses the power of large language models (LLMs) to generate highly accurate and natural-sounding speech.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work