VoxCPM TTS - Zero Shot Text to Speech

How to Use the VoxCPM TTS Workflow

➡️ Option 1: With Audio (Voice Cloning)

This option lets you clone a voice from an audio file.

1. Add your audio: Connect a .wav file of the voice you want to clone to the LoadAudio node.
2. Provide the transcript: In the text box located under the retry_threshold node, paste the exact transcript of the provided audio. This helps the model accurately clone the voice.
3. Enter your desired text: Type the text you want the model to generate into the top text box.

➡️ Option 2: Without Audio (Standard TTS)

This option provides standard text-to-speech without voice cloning.

1. Bypass the audio node: Disconnect or bypass the LoadAudio node.
2. Clear the transcript: Make sure there is no text in the text box under the retry_threshold node.
3. Enter your desired text: Type the text you want to generate into the top text box.