How to Use Gemini TTS API — Complete Tutorial 2026

pip install requests

import requests
import os

# Get your API key from RapidAPI: https://rapidapi.com/nexaquency/api/gemini-tts
RAPIDAPI_KEY = os.environ.get("RAPIDAPI_KEY", "your-rapidapi-key-here")
API_HOST = "gemini-tts.p.rapidapi.com"

def text_to_speech(text: str, voice_id: str = "default", **kwargs) -> bytes:
    """
    Convert text to speech using Gemini TTS API via NexaAPI on RapidAPI.
    
    Args:
        text: Text to convert to speech
        voice_id: Voice identifier (default: "default")
        **kwargs: Additional parameters (stability, similarity_boost, etc.)
    
    Returns:
        Audio bytes (MP3 format)
    """
    url = f"https://{API_HOST}/synthesize"
    
    headers = {
        "x-rapidapi-key": RAPIDAPI_KEY,
        "x-rapidapi-host": API_HOST,
        "Content-Type": "application/json"
    }
    
    payload = {
        "text": text,
        "voice_id": voice_id,
        **kwargs
    }
    
    response = requests.post(url, json=payload, headers=headers)
    response.raise_for_status()
    
    # Return audio bytes or URL depending on API response
    content_type = response.headers.get("content-type", "")
    if "audio" in content_type:
        return response.content
    else:
        result = response.json()
        return result.get("audio_url") or result.get("url")


def save_audio(text: str, output_path: str, voice_id: str = "default") -> str:
    """Generate speech and save to file."""
    audio_data = text_to_speech(text, voice_id)
    
    if isinstance(audio_data, bytes):
        with open(output_path, "wb") as f:
            f.write(audio_data)
        print(f"Audio saved to: {output_path}")
    else:
        print(f"Audio URL: {audio_data}")
    
    return output_path


if __name__ == "__main__":
    # Example 1: Basic TTS
    print("Converting text to speech...")
    result = text_to_speech(
        text="Welcome to NexaAPI. The fastest and most affordable AI API platform in 2026.",
        voice_id="default"
    )
    print(f"Result: {result if isinstance(result, str) else f'{len(result)} bytes of audio'}")
    
    # Example 2: Save to file
    save_audio(
        text="This is a longer example of text-to-speech generation using Gemini TTS via NexaAPI.",
        output_path="output_speech.mp3"
    )

import requests
import time
import os

RAPIDAPI_KEY = os.environ.get("RAPIDAPI_KEY")
API_HOST = "gemini-tts.p.rapidapi.com"

def api_call_with_retry(endpoint: str, payload: dict, max_retries: int = 3) -> dict:
    """Make API call with exponential backoff retry."""
    for attempt in range(max_retries):
        try:
            response = requests.post(
                f"https://{API_HOST}/{endpoint}",
                json=payload,
                headers={
                    "x-rapidapi-key": RAPIDAPI_KEY,
                    "x-rapidapi-host": API_HOST,
                    "Content-Type": "application/json"
                },
                timeout=120  # 2 minute timeout for generation
            )
            response.raise_for_status()
            return response.json()
        except requests.exceptions.HTTPError as e:
            if e.response.status_code == 429:
                wait = 2 ** attempt
                print(f"Rate limited. Waiting {wait}s...")
                time.sleep(wait)
            elif e.response.status_code >= 500:
                print(f"Server error. Retry {attempt+1}/{max_retries}")
                time.sleep(2)
            else:
                raise
        except requests.exceptions.Timeout:
            print(f"Timeout. Retry {attempt+1}/{max_retries}")
            time.sleep(5)
    raise Exception(f"Failed after {max_retries} retries")