Unlocking the Power of Voice: A Comprehensive Guide to Using OpenAI's Text-to-Speech API

Unreal Speech

• 2 min read
Unlocking the Power of Voice: A Comprehensive Guide to Using OpenAI's Text-to-Speech API

OpenAI's Text-to-Speech (TTS) API is a technology that transforms written text into spoken words, providing a natural-sounding voice output. The API offers two model variations:

  1. TTS-1: Optimized for real-time text-to-speech applications.
  2. TTS-1-HD: Focused on high-quality audio output.

This API comes with six prebuilt voices and can be used for narrating blog posts, producing multilingual audio, and offering real-time audio output. Users must disclose that the voice heard is AI-generated.

Getting Started with the TTS API:

  • Prerequisites: An OpenAI account with funding, Python 3.7 or newer, and an Integrated Development Environment (IDE).
  • Step 1: Generate an API Key: Log in to your OpenAI account, navigate to the API Keys section, and create a new secret key.
  • Step 2: Create a Virtual Environment: This isolates your Python project and its dependencies.
from pathlib import Path
from openai import OpenAI

client = OpenAI()

speech_file_path = Path(__file__).parent / "speech.mp3"
response = client.audio.speech.create(
  model="tts-1",
  voice="alloy",
  input="Today is a wonderful day to build something people love!"
)

response.stream_to_file(speech_file_path)

Bad Practice: Directly adding the key to the OpenAI object.

Good Practice: Using dotenv for key management.


pip install python-dotenv

Use Environment Variables:

import os
from pathlib import Path
from openai import OpenAI
from dotenv import load_dotenv

load_dotenv() 
SECRET_KEY = os.getenv("SECRET_KEY")

client = OpenAI(api_key=SECRET_KEY)

speech_file_path = Path(__file__).parent / "speech.mp3"
response = client.audio.speech.create(
  model="tts-1",
  voice="alloy",
  input="Today is a wonderful day to build something people love!"
)

response.stream_to_file(speech_file_path)

Customizing Voice and Output:

You can choose from six voices (Alloy, Echo, Fable, Onyx, Nova, Shimmer) and various output formats (MP3, AAC, FLAC, Opus).

Real-World Applications:

Useful for narrating content, creating multilingual audio, enhancing real-time interactions in games or with virtual assistants.

API Limits and Pricing:

  • 50 RPM for paid accounts.
  • Max input: 4,096 characters.
  • Pricing varies between standard and HD models.

In summary, OpenAI's TTS API is a versatile tool for text-to-speech conversion, offering customization and suitable for a range of applications, with clear usage limits and cost guidelines.