OpenAI FM - Open-Source Text-to-Speech Demo

VISIT OPENAI FM

OpenAI FM Brief Overview

OpenAI FM (also known as OpenAI.fm) is an open-source web demo created to showcase OpenAI’s text-to-speech capabilities through the OpenAI Speech API. Instead of being a standalone “AI voice generator app” with its own hosted subscription, it’s essentially a ready-made Next.js project you can run locally (or deploy yourself) to experiment with speech generation.

The experience is designed to be simple: you pick a voice, choose a style “vibe” (a preset that influences how the voice should sound), paste or edit a script, and generate audio you can play back immediately. It’s a practical reference for developers who want to understand what a clean text-to-speech testing interface looks like and how a basic Speech API workflow can be wrapped into a friendly UI.

Simple How-to-Use

1. Get an OpenAI API key
   • Create an OpenAI account and generate an API key from your dashboard.
2. Clone the repository
   • git clone https://github.com/openai/openai-fm.git
3. Add your API key
   • Either set OPENAI_API_KEY as a system environment variable, or
   • Create a .env file in the project root and add:
     OPENAI_API_KEY=your_key_here
4. Install and run
   • In the project directory:
     npm install
npm run dev
5. Use the demo UI
   • Open http://localhost:3000
   • Select a voice
   • Select a vibe (style preset)
   • Edit the script
   • Click Play to generate speech, then optionally Download the audio
6. Optional: enable sharing
   • If you want the app’s sharing feature, connect a hosted Postgres database and set POSTGRES_URL in your .env.

Key Features and Functions

• Voice selection: Quickly switch between multiple built-in voice options to compare tone and delivery.
• “Vibe” presets: Choose from style/personality presets (for example, different narrative tones or character-like deliveries) to guide how the speech should sound.
• Script editor: Paste or write text directly in the interface before generating audio.
• One-click playback: Generate speech audio and play it back immediately from the UI.
• Download audio: Export generated speech as an audio file for use in videos, prototypes, demos, or accessibility tooling.
• Share capability (optional): A sharing workflow can be enabled by connecting a Postgres database.
• Self-hostable Next.js codebase: Useful as a learning reference or as a starting point for a custom internal voice playground.

OpenAI FM Pricing

• OpenAI FM itself is free: The repository is open-source, so there’s no license fee to download, run, or modify the demo.
• You pay for API usage: Speech generation calls the OpenAI API, so costs depend on the speech model and how much text you convert to audio.
  • OpenAI lists TTS speech generation at $15.00 per 1M characters
  • TTS HD is listed at $30.00 per 1M characters
• Example cost (simple estimate):
  • If you generate 100,000 characters of speech:
    • Standard TTS: 100,000 / 1,000,000 = 0.1 → 0.1 × $15.00 = $1.50
    • TTS HD: 0.1 × $30.00 = $3.00
• Self-hosting costs (optional): If you deploy publicly, you’re responsible for any usage billed to your API key. If you enable sharing, you may also pay for hosted Postgres and hosting infrastructure.

Other Popular AI Tools

ChatCSV – AI Data Analyst

FormWise -White Label AI Tools

FreeImage AI – AI Image Generator

TLDR This -AI Text Summarizer Tool

Sendsteps AI – AI Presentation Maker