🎙️ ComfyUI-Gemini_TTS

A powerful ComfyUI custom node that brings Google’s Gemini TTS capabilities directly to your workflow. Generate high-quality speech with 30+ voices supporting both free and paid tiers.

✨ Features

30+ Premium Voices: Male and female voices with unique characteristics

Dual Tier Support: Free tier with generous limits + Paid tier for production use

Smart Fallback: Automatic model switching when quotas are reached

Voice Characteristics: Detailed voice info with personality descriptions

Flexible Configuration: Environment variables, node parameters, or config file

Robust Error Handling: Clear error messages and automatic retry logic

Real-time Pricing: Cost estimates for paid tier usage

🚀 Quick Start

1. Installation

Clone or download this repository to your ComfyUI custom nodes folder:

“`bash

cd ComfyUI/custom_nodes/

git clone https://github.com/ShmuelRonen/ComfyUI-Gemini_TTS.git

“`

Install dependencies:

“`bash

cd gemini-tts-node

pip install google-generativeai requests torch torchaudio numpy

“`

Restart ComfyUI – The node will appear as “🎙️ Gemini Text-to-Speech”

2. Get Your API Key

Free Tier (Recommended to Start)

Go to Google AI Studio

Sign in with your Google account

Click “Get API Key” → “Create API Key”

Select “Create API key in new project”

Copy your API key (starts with AIza...)

Paid Tier (For Production)

See the Paid Tier Setup section below.

3. Configure the Node

Option A: Environment Variable (Recommended)

export GEMINI_API_KEY="your_api_key_here"

Option B: Direct Input

Enter your API key directly in the node’s api_key field

The node will save it automatically for future use

🎭 Available Voices

Female Voices (14 total)

Aoede – Breezy and natural

Kore – Firm and confident

Leda – Youthful and energetic

Zephyr – Bright and cheerful

Autonoe – Bright and optimistic

Callirhoe – Easy-going and relaxed

Despina – Smooth and flowing

Erinome – Clear and precise

Gacrux – Mature and experienced

Laomedeia – Upbeat and lively

Pulcherrima – Forward and expressive

Sulafat – Warm and welcoming

Vindemiatrix – Gentle and kind

Achernar – Soft and gentle

Male Voices (16 total)

Puck – Upbeat and energetic (default)

Charon – Informative and clear

Fenrir – Excitable and dynamic

Orus – Firm and decisive

Achird – Friendly and approachable

Algenib – Gravelly texture

Algieba – Smooth and pleasant

Alnilam – Firm and strong

Enceladus – Breathy and soft

Iapetus – Clear and articulate

Rasalgethi – Informative and professional

Sadachbia – Lively and animated

Sadaltager – Knowledgeable and authoritative

Schedar – Even and balanced

Umbriel – Easy-going and calm

Zubenelgenubi – Casual and conversational

⚙️ Node Parameters

Required Parameters

prompt: Text to convert to speech (supports “Say:” prefix)

tts_model: Choose between:

gemini-2.5-pro-preview-tts (Higher quality, slower)

gemini-2.5-flash-preview-tts (Faster, good quality)

voice: Select from 30+ available voices

temperature: Control creativity (0.0-2.0, default: 1.0)

Optional Parameters

api_key: Enter API key directly (auto-saved)

auto_fallback_to_flash: Auto-switch to Flash if Pro is rate-limited

retry_delay: Wait time between retries (10-120 seconds)

use_paid_tier: Enable paid billing for higher quotas

billing_project_id: Google Cloud project ID for billing

aggressive_retry: More retry attempts for better reliability

show_voice_info: Display voice characteristics in output

💰 Paid Tier Setup

Why Upgrade to Paid Tier?

| Feature | Free Tier | Paid Tier |

|———|———–|———–|

| Quota Limits | Low (good for testing) | High (production ready) |

| Rate Limits | Restrictive | Generous |

| Priority Access | Standard | Premium |

| Cost | Free | ~$0.001-0.02 per request |

Step-by-Step Paid Setup

1. Create Google Cloud Project

Go to Google Cloud Console

Click “New Project” or select existing project

Enter project name (e.g., “my-gemini-tts”)

Note your Project ID (not the name – this is important!)

2. Enable Billing

In Google Cloud Console, go to Billing

Click “Link a billing account” or “Enable billing”

Add a payment method (credit card required)

Verify billing is active on your project

3. Enable the Gemini API

Go to APIs & Services > Library

Search for “Generative Language API”

Click “Enable” on the Generative Language API

Wait for activation (usually instant)

4. Create API Key

Go to APIs & Services > Credentials

Click “Create Credentials” > “API Key”

Copy your new API key

Optional: Restrict the key to “Generative Language API” for security

5. Configure the Node

Set these parameters in the node:

use_paid_tier: True

billing_project_id: Your Project ID from step 1

api_key: Your API key from step 4

💵 Pricing Information

Gemini 2.5 Pro TTS:

Input: $1.00 per 1M tokens

Output: $20.00 per 1M tokens

~$0.01-0.02 per typical request

Gemini 2.5 Flash TTS:

Input: $0.50 per 1M tokens

Output: $10.00 per 1M tokens

~$0.005-0.01 per typical request

*Typical 20-word sentence costs less than $0.02*

🔧 Troubleshooting

Common Issues

“API key not valid” Error

Solution: Verify your API key starts with AIza and is ~39 characters

Check: API key hasn’t expired or been deleted

Verify: You’re using the correct key from Google AI Studio or Cloud Console

“Rate limit exceeded” Error

Free Tier: Wait 60 seconds or try Flash model

Solution: Enable paid tier for higher quotas

Temporary: Use auto_fallback_to_flash = True

“Billing project not found” Error

Check: Use Project ID, not project name

Verify: Project exists and billing is enabled

Confirm: API key belongs to the same project

“Permission denied” Error

Verify: Generative Language API is enabled

Check: API key has proper permissions

Ensure: Billing is active if using paid tier

Configuration Files

The node creates a config.json file to save your settings:

{
    "GEMINI_API_KEY": "your_key_here",
    "use_paid_tier": true,
    "billing_project_id": "your-project-id"
}

Debug Information

Enable debugging by checking console output:

Green ✅: Successful operations

Yellow ⚠️: Warnings and fallbacks

Red ❌: Errors requiring attention

📝 Usage Examples

Basic Text-to-Speech

Prompt: "Hello, welcome to our presentation today."
Model: gemini-2.5-flash-preview-tts
Voice: [F] Zephyr
Temperature: 1.0

Expressive Reading

Prompt: "Say: Once upon a time, in a land far, far away..."
Model: gemini-2.5-pro-preview-tts  
Voice: [M] Charon
Temperature: 1.5
Show Voice Info: True

Production Setup

Use Paid Tier: True
Billing Project ID: my-production-project-123
Aggressive Retry: True
Model: gemini-2.5-pro-preview-tts

🛡️ Security Best Practices

Protect Your API Key: Never commit API keys to version control

Use Environment Variables: Set GEMINI_API_KEY in your environment

Restrict API Keys: Limit to specific APIs in Google Cloud Console

Monitor Usage: Check Google Cloud billing dashboard regularly

Project Isolation: Use separate projects for development vs production

🔄 Updates and Compatibility

ComfyUI: Compatible with latest versions

Python: Requires Python 3.8+

Dependencies: Auto-updated through pip

Voice Library: Automatically synced with Google’s latest voices

📞 Support

Common Solutions

Restart ComfyUI after installation or configuration changes

Check Console Output for detailed error messages

Verify API Key Format (should start with AIza)

Confirm Project Settings in Google Cloud Console

Getting Help

Check the troubleshooting section above

Review console output for specific error messages

Verify your Google Cloud project configuration

Ensure billing is properly enabled for paid tier

📜 License

This project is provided as-is for educational and commercial use. Google Gemini API usage is subject to Google’s terms of service and pricing.

---

🎉 Ready to generate amazing speech with Gemini TTS!

*Last updated: May 2025*