AI Text to Speech Voiceover - MP3, WAV, Opus and AAC
Have your AI turn a script into a voiceover you can download. Runs on your own OpenAI key.
Paste in a script, or a list of them, and each one comes back as an audio file read by an AI voice: MP3 unless you ask for WAV, Opus or AAC. The speech is made on your own OpenAI key, so OpenAI bills you for it directly and this tool charges per finished voiceover. Leave the key out and you get a free demo record instead of audio.
Who uses an AI voiceover tool in batches
- Channels that publish narrated videos. A week of scripts goes in as one list, and the files come back in the same voice at the same speed, numbered in the order you gave them.
- Courses and training material. When one lesson changes, you re-voice that lesson's script and nothing else, and it still sounds like the rest.
- Long reads. An article of several thousand words is split into parts behind the scenes and joined back into one file, so nobody has to cut it up by hand.
A worked example
Say you have twelve product videos with a script of about 150 words each. Put the twelve scripts in texts, set voice to nova and model to tts-1-hd, and leave format on MP3.
The run writes twelve MP3 files to its storage and twelve rows, one per script, in your order. Each row links to its file and says how long it runs and how many characters were read. That's twelve voiceovers on this tool's bill, and OpenAI charges your key for the characters at its own rates.
A script that fails gets a line in a notices record in the run's storage instead of a row, and isn't charged. A key OpenAI refuses stops the batch there, so one typo doesn't run down your whole list.
What a row looks like
| Field | What it tells you |
|---|---|
index | Which script this is, counting from 1 in the order you gave them |
audioUrl | A direct link to the finished file |
durationSeconds | How long the audio runs. A 0 means the file couldn't be measured, and the audio is still there |
characters | How long the script was |
chunks | How many parts a long script was split into before joining |
voice, model, format | What it was made with |
textPreview | The first 160 characters, to tell the rows apart |
When it's the wrong tool
- You want your own voice, or a clone of someone's. It has OpenAI's six voices and nothing else.
- You need pauses, emphasis or pronunciation marked up. There's no SSML. Punctuation and
speedare the only controls. - You already have the video and want it in another language. AI Video Dubber re-voices it.
- You want the other direction, speech into text. That's the Video & Audio Transcriber.
- You want the voice laid over pictures. Take the
audioUrlto Storyboard Video Generator. - It is not an official tool of any platform or company named on this page. Dami's Studio is independent and is not affiliated with or endorsed by any of them.
How it works
- 1Open it on Apify
The Open on Apify button takes you there. Nothing to install.
- 2Set the inputs
The form opens with an example in Text / script. Put in your own and check the other settings.
- 3Click Run
It runs on Apify's servers, on your own Apify account.
- 4Export the results
Download them from the run, or read them over Apify's API.
Pricing
$0.04 per voiceover
Live price on Apify's Free plan, read on 5 November 2026. Paid plans pay a little less. You are billed on your own Apify account. The full price list is on the Apify page.
Use it from your AI chat
Add Apify's connector to your AI chat once (setup steps), then paste this:
Use the AI Text to Speech Voiceover (dami_studio/ai-tts-voiceover) to turn this script into a voiceover: "Welcome back to the channel. Today, we're diving into one of the strangest mysteries of the deep ocean.", and return one audio file per script, and a row with its link, length, voice and character count.This tool runs on your own OpenAI API key. Never paste a key into a chat: run it from the Apify page, where the key goes in the input form.
Some apps ask for a server address instead. This one loads just this tool:
https://mcp.apify.com?tools=dami_studio/ai-tts-voiceoverEvery run your AI chat starts goes on your own Apify account, at the prices on this tool's Apify page.
Questions people ask
Can I get text to speech as an MP3?
Yes. MP3 is what you get unless you set format to WAV, Opus or AAC.
Which AI voices can I choose from?
Alloy, echo, fable, onyx, nova and shimmer. Onyx is used if you don't pick one. For the model, tts-1-hd sounds cleaner than the default tts-1, and OpenAI charges your key more for it.
How long can a script be?
There's no limit per script. Long ones are split and joined back, and the run's own time limit is the real ceiling: the run stops shortly before it and keeps the files it has finished.
Is there a text to speech API for bulk jobs?
This is one. It runs from Apify's API like the other tools there, so a script of your own can send a list of texts and collect the files. The API tab on its Apify page has the calls written out.
How much does AI Text to Speech Voiceover cost?
$0.04 per voiceover on Apify's Free plan. Paid plans pay a little less. It is billed on your own Apify account.
Can I run AI Text to Speech Voiceover from my AI chat?
Yes, from any app that supports MCP connectors. Add Apify's connector (https://mcp.apify.com/), sign in with your Apify account, and ask for dami_studio/ai-tts-voiceover. The steps are on the Use with AI page. Each run goes on your own Apify account, at the prices on its Apify page.
Do I need to write code to use AI Text to Speech Voiceover?
No. Open it on Apify, fill in the form, click Run and download the results.