Lazybird

By Lazybird

Overview

Turn scripts into ready-to-use voiceover MP3s without leaving ChatGPT. Write or polish your script with ChatGPT, then ask Lazybird to read it aloud: you get an audio player in the chat and an MP3 you can download and drop into your video, podcast, ad, audiobook or course. What you can do: - Find the right voice by describing it ("a calm British narrator for a nature documentary"). Lazybird searches 500+ voices in 150+ languages and accents, plus the voices in your own library, including your cloned voices. - Generate speech from any text with the voice you picked, adjusting the speaking speed or the language and accent. - Design a brand-new voice from a description ("a raspy, ancient wizard with a slow, mysterious delivery"), listen to the sample, and save it to your Lazybird library to reuse in ChatGPT or in Lazybird Studio. - Check how many characters you have left. Connect your Lazybird account the first time you use it. Lazybird can't see your password, payment details or API keys, and you can disconnect ChatGPT any time in Lazybird Settings → Connected apps.

Tools

design_voice

ChatGPT

Create a brand-new AI voice from a description (age, gender, accent, tone, energy), e.g. for a character, a brand or a narrator, and hear it say a sample line. Uses characters from the user's balance (at least 50). The voice isn't saved; to keep it, call save_voice with the returned sampleUrl.

generate_speech

ChatGPT

Turn a script into a ready-to-use voiceover: speaks the text in a Lazybird voice and returns a link to the MP3, which the user can play, download or put in a video, podcast, ad or course. Uses one character from the user's balance per character of text. Up to 5000 characters per call; split longer scripts into parts.

get_account

ChatGPT

The connected Lazybird account: email, plan name, and characters used, left and when the allowance resets.

save_voice

ChatGPT

Add a voice to the user's Lazybird account from an audio sample: a sample returned by design_voice, or a public link to a clear 10-60 second recording of one speaker (MP3, WAV or OGG) to clone. Only clone a real person's voice when it's the user's own voice or the speaker has given permission; set rightsConfirmed only after the user confirms that. Takes one of the account's voice slots.

search_voices

ChatGPT

Find the right voice for a voiceover: the catalog of 500+ AI voices in 150+ languages and accents, plus the user's own cloned or designed voices and the ones their team shares. Filter by words (name, accent, style, use case such as narrator or ad), language and gender. Returns voice ids for generate_speech, with preview audio links the user can listen to.