ChatTTS Voice Generator

AI-optimized conversational text-to-speech using a local ChatTTS server. Designed to generate highly natural human voices with realistic laughs, breathing, and pauses.

Generate Speech

CODEBLOCK0

Options

• - --seed <number>: Random seed to fix the speaker's voice/timbre (default: 2048). Change this number to switch between different male/female voices.
• INLINECODE1: Controls the emotional variance and stability (default: 0.3). Lower is more stable and clear, higher (e.g., 0.6) is more expressive but might mumble.
• INLINECODE2: Top P sampling parameter for voice generation (default: 0.7).

Notes

• - Requires the local ChatTTS FastAPI server to be running (default target: http://172.23.252.114:8020).
• Ensure CHATTTSAPIURL is set in your .env file if the API is hosted on a different machine.
• PRO TIP: Always try to insert [laugh] (laughter) and [uv_break] (pauses/breaths) into the text to make the generated voice sound exactly like a real human.
• The script will return the absolute local file path of the generated .wav audio file.