Generate realistic AI voiceovers directly inside ComfyUI using this Kokoro TTS Workflow. This setup allows you to create free, offline text-to-speech audio, blending multiple speaker models for custom voices. Ideal for narration, storytelling, and character dialogue generation—all without cloud services.

---

🎯 Features

• Free & Local – No online services required; fully runs offline using Kokoro.
• Multiple Speaker Voices – Includes support for models like am_onyx and am_adam.
• Voice Blending – Combine two voices via KokoroSpeakerCombiner for unique speech tone.
• Text-to-Speech Node – KokoroGenerator turns your typed text into audio using the selected voice.
• Audio Export – Output is saved as a playable .wav file via SaveAudio.

---

💡 Use Cases

• YouTube Narration – Generate commentary and explanations for videos.
• Game Development – Create character dialogue with distinct tones.
• Voiceover for Animation – Narrate animated scenes or intros locally.
• Storytelling & Audiobooks – Read out long texts with expressive voice control.
• Virtual Assistants – Add speech to AI bots or desktop assistants.

---

⚙️ How It Works

1. Load Speakers – Use two KokoroSpeaker nodes to select different voices (e.g. am_onyx, am_adam).
2. Blend Voices (Optional) – Use KokoroSpeakerCombiner to merge voices with adjustable ratio.
3. Type Your Text – Input your line into KokoroGenerator, choose speed/language settings.
4. Generate Speech – Connect the combined speaker into KokoroGenerator to create the voice audio.
5. Save Audio – Use SaveAudio to export the voiceover as a .wav file.

---

Credits: pixaroma