- What voices are available?
- Six voices are available through the OpenAI TTS API: Alloy (neutral, versatile), Echo (male, steady), Fable (British male, expressive), Onyx (male, deep and authoritative), Nova (female, warm) and Shimmer (female, gentle). All six are natural-sounding and suitable for professional use.
- What is the maximum text length?
- The tool accepts up to 4,000 characters per conversion. For longer articles, split the content into sections - introduction, body, conclusion - and convert each separately, then join the MP3 files in any audio editor.
- What format is the output file?
- The output is an MP3 file (audio/mpeg). MP3 is compatible with all audio players, podcast platforms, video editors and web browsers. It can be uploaded directly to platforms like Spotify, Apple Podcasts or embedded in a website using an HTML audio element.
- Is my text stored or used to train the AI?
- No. Your text is sent to OpenAI's API to generate the audio and is not retained by this tool. JohnB.io does not log or store the content you submit. See the OpenAI Privacy Policy for full details.
- What does the daily usage limit mean?
- There is a per-IP daily limit to keep the tool available for everyone without requiring a login. If you hit the limit, it resets at midnight UTC. For higher-volume use, connect via LinkedIn.
- How much does OpenAI TTS cost?
- The tts-1 model costs $0.015 per 1,000 characters of input text. A typical 500-word article (around 3,000 characters) costs approximately $0.045 to generate. The tts-1-hd model at $0.030 per 1,000 characters offers marginally better quality for studio-grade production use.