Mix It Up text to speech: voices, limits and character voice lines
Mix It Up is a free streaming bot, and its Text To Speech action can read chat messages, redeems and alerts aloud. Which voices you get depends on the provider you pick, and some providers have rate limits that explain many “TTS not working” moments.
This guide sums up the providers from the Mix It Up docs and shows how to add lines in game, cartoon or your own voice with the Sound action.
Providers in the Text To Speech action
According to the Mix It Up docs, the Text To Speech action can use Windows Text To Speech, ResponsiveVoice, TTS.Monster through its API or an Overlay URL, Amazon Polly, Microsoft Azure Speech, TikTok TTS, Edge TTS and Google Cloud TTS. Windows Text To Speech uses the voices installed on your PC, though not every installed voice can generate audio from text.
Why TTS seems not to work
Several providers are limited. The docs list a rate limit of 3 actions per minute for TikTok TTS and for Edge TTS, and a default cooldown of one Text To Speech use every 5 minutes for Amazon Polly and Microsoft Azure Speech, which you can lift by connecting your own account. ResponsiveVoice only works through the Mix It Up Overlay; without it you get an error asking you to enable the Overlay in Services.
Character voices with the Sound action
For fixed lines, such as a follow alert, a channel point redeem or a raid welcome, you do not need live TTS at all. Voice each line in advance in a character voice and play it with the Sound action, which plays a sound file at a volume from 0 to 100 on the output device you choose. Hear voices free on their pages with a line of 30 to 100 characters, no account needed, and download the recording.
Make the lines offline
For a whole set of alerts, buy a voice once, download its file and install the free engine with python -m pip install superclonica-cpu. Then voice each line: superclonica --text "Thanks for the follow, welcome to the crew!" --voice voice.json --lang en --normalize --output follow.mp3. MP3 is what the Mix It Up docs suggest when a file will not play; --normalize keeps every alert at the same loudness. It runs offline on an ordinary processor, with no character limit.
Mix both
Use a built-in provider for reading arbitrary chat messages, and pre-voiced character lines for the moments that repeat. Viewers hear a familiar voice for every follow or redeem, and the lines never hit a rate limit. Character voices in the catalog are fan-made, for fan works, parody and personal projects; for a voice of your own, train it in the studio from a 5 to 10 second recording.