ComfyUI-SparkTTS

ComfyUI-SparkTTS
★ 125

文本转语音语音克隆音频处理自定义语音
ComfyUI-SparkTTS 是 ComfyUI 的 SparkTTS 文本转语音节点，基于大模型生成高保真自然语音，支持自定义语音、语音克隆、音频处理与录音。
💡 在 ComfyUI 流程中生成或克隆自然语音并进行音频处理
🍴 13 Forks💻 Python🔄 2025-04-15
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/9b747dd328b1
📦 requirements.txt
einops>=0.8.1
einx>=0.3.0
omegaconf>=2.3.0
soundfile>=0.12.1
torch>=2.0.0
transformers>=4.30.0
huggingface_hub>=0.19.0
torchaudio>=0.13.0
librosa>=0.10.0
sounddevice>=0.4.6
scipy>=1.10.0
numpy>=1.24.0
soxr>=0.3.5
SparkTTS_Nodes
kokoroTTS_menu
📄 README

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.

News & Updates

  • 2025/03/21: Update ComfyUI-SparkTTS to v1.1.0 ( update.md )

  • Integrated internationalization (i18n) support for multiple languages.
  • Improved user interface for dynamic language switching.
  • Enhanced accessibility for non-English speaking users with fully translatable features.

    • Features

    ComfyUI-SparkTTS provides the following main functionalities:

  • Voice Creation: Create a customized voice by adjusting parameters like gender, pitch, and speed.
  • Voice Cloning: Clone a voice from a reference audio sample.
  • Advanced Voice Cloning: Clone a voice from a reference audio with control over pitch and speed.
  • Audio Processing: Load and process audio files.
  • Audio Recording: Directly record audio for voice cloning or processing.

    • Installation

    Method 1. install on ComfyUI-Manager, search Comfyui-SparkTTS and install

    install requirment.txt in the ComfyUI-SparkTTS folder

    “`bash

    ./ComfyUI/python_embeded/python -m pip install -r requirements.txt

    “`

    Method 2. Clone this repository to your ComfyUI custom_nodes folder:

    “`bash

    cd ComfyUI/custom_nodes

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

    “`

    install requirment.txt in the ComfyUI-SparkTTS folder

    “`bash

    ./ComfyUI/python_embeded/python -m pip install -r requirements.txt

    “`

    Method 3: Install via Comfy CLI

    Ensure pip install comfy-cli is installed.

    Installing ComfyUI comfy install (if you don’t have ComfyUI Installed)

    install the ComfyUI-SparkTTS, use the following command:

    “`bash

    comfy node registry-install Comfyui-Spark-TTS

    “`

    install requirment.txt in the ComfyUI-SparkTTS folder

    “`bash

    ./ComfyUI/python_embeded/python -m pip install -r requirements.txt

    “`

    4. Manually download the models:

  • The model will be automatically downloaded to ComfyUI/models/TTS/SparkTTS/ when first time using the custom node.
  • Manually download the SparkTTS-2.0 model by visiting this link, then download the files and place them in the ComfyUI/models/TTS/SparkTTS/ folder.

    • Nodes

    SparkTTS Voice Creator 🔊

    This node allows you to create a customized voice by adjusting parameters.

    Inputs:

  • text: Text to synthesize.
  • gender: Gender of the voice (female or male).
  • pitch: Pitch level of the voice (very_low, low, moderate, high, very_high).
  • speed: Speed level of the voice (very_low, low, moderate, high, very_high).
  • batch_texts (optional): Additional texts for better control over pacing and intonation.

    • Outputs:

  • audio: Generated audio with the customized voice.

    • SparkTTS Voice Clone 🔊

    This node allows you to clone a voice from a reference audio sample.

    Inputs:

  • text: Text to synthesize with the cloned voice.
  • reference_audio: The audio sample to clone the voice from.
  • reference_text: Transcript of the reference audio to improve cloning quality.
  • max_tokens: Controls the maximum length of generated speech.
  • batch_texts (optional): Additional texts for better control over pacing and intonation.

    • Outputs:

  • audio: Generated audio with the cloned voice.

    • SparkTTS Advanced Voice Clone 🔊

    This node allows you to clone a voice from a reference audio with control over pitch and speed.

    Inputs:

  • text: Text to synthesize with the cloned voice.
  • reference_audio: The audio sample to clone the voice from.
  • reference_text: Transcript of the reference audio to improve cloning quality.
  • pitch: Pitch level of the voice.
  • speed: Speed level of the voice.
  • max_tokens: Controls the maximum length of generated speech.
  • batch_texts (optional): Additional texts for better control over pacing and intonation.

    • Outputs:

  • audio: Generated audio with the cloned voice.

    • Audio Recorder 🔊

    This node allows you to directly record audio.

    Inputs:

  • recording: Set to True to start recording audio.
  • recording_duration: Recording duration in seconds.
  • sample_rate: Audio sample rate.
  • noise_threshold: Noise reduction threshold.
  • smoothing_kernel_size: Size of the kernel used for smoothing the audio signal.

    • Outputs:

  • audio: Recorded audio data.

    • Example Workflows

    Check the example_workflows directory for example workflows.

    Supported Languages

    SparkTTS currently supports the following languages:

  • English
  • Chinese

    • License

    GPL-3.0 License