This guide will show you how to generate your first audio clip with Rime’s TTS API and experiment with different voices and speech customizations.

​
Prerequisites

For this guide you will need a Rime API token. To get a token, create a free Rime account. Then, from the Rime dashboard, navigate to the API Tokens page and copy the API key for later use. We will be using Python to send requests to the Rime API so install Python 3.10 or later.

​
Set Up Your Environment

Create a virtual environment to keep your project dependencies isolated. 
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
Install the requests library: 
pip install requests

​
Create Your Python Script

Create a file called rime_hello_world.py and import the requests library. 
import requests
Next, create the request to the Rime API. Set the request headers specifying the Rime API key that you copied. 
RIME_API_KEY = "your_api_key_here"

headers = {
    "Accept": "audio/mp3",
    "Authorization": f"Bearer {RIME_API_KEY}",
    "Content-Type": "application/json"
}
Now, create a payload where we will specify the details of the request to the API. 
payload = {
    "text": "Hello! This is Rime speaking.",
    "speaker": "celeste",
    "modelId": "arcana"
}
There are many optional parameters that we could send to the API but the following are required:
  • text - The text to convert to speech
  • speaker - The voice to use (see voices)
  • modelId - Use arcana for the most realistic voices, or mistv2 for faster synthesis
See the api-reference for more optional parameters. Now that our headers and payload are created, make a POST request to the Rime API and write the streamed audio response to a file. 
with requests.post(
    "https://users.rime.ai/v1/rime-tts",
    headers=headers,
    json=payload,
    stream=True
) as response:
    response.raise_for_status()
    
    with open("output.mp3", "wb") as f:
        for chunk in response.iter_content(chunk_size=4096):
            if chunk:
                f.write(chunk)

print("Audio saved to output.mp3")
Streaming writes audio chunks as they arrive, reducing memory usage for longer text. Test the script by simply running it from the terminal. 
python rime_hello_world.py
On a successful run, we should see confirmation that our audio file is saved. 
'Audio saved to output.mp3'

​
Choose a Voice

Rime offers a range of voices with different personalities. To change the voice, update the speaker parameter in your request. 
json={
    "text": "Hello! This is Rime speaking.",
    "speaker": "orion",  # Try different voices here
    "modelId": "arcana",
    "samplingRate": 24000
}
Browse all available voices on the voices page.

​
Custom Pronunciation

The mistv2 model offers a way to specify pronunciation of brand names or uncommon words using Rime’s phonetic alphabet. Add the custom pronunciation in curly brackets and set phonemizeBetweenBrackets to true: 
json={
    "text": "Welcome to {r1Ym} labs.",
    "speaker": "peak",
    "modelId": "mistv2",
    "samplingRate": 24000,
    "phonemizeBetweenBrackets": True
}
Use the Pronunciation tool in the dashboard to generate phonetic strings for any word.

​
Next Steps

Now that you can generate TTS audio, the next guide will explore how to enable a real-time conversation between you and an agent. Before that, check out these resources to get more familiar with Rime:
  • Models - Compare Arcana (realistic) vs Mist v2 (fast)