Izabela

Your speech assistant ✨

What is Izabela?

Izabela is a text-to-speech application that allows you to communicate with an artificial voice on your speakers or as an audio input!

It was designed primarily to help mute people and people with speech disabilities communicate with their friends over applications such as Discord or games that don't include text chat without interrupting their activity. Over time, it evolved to be a tool for anyone who wants to communicate with text-to-speech and for that reason, you can see Izabela as a sort of speech assistant!

Features

• Appears above any non-fullscreen application
• Multiple text-to-speech engines (Amazon Polly, Microsoft Azure, Google Cloud, IBM Watson, etc..) - see list
• Multiple voices (per engine)
• Supports custom text-to-speech engines (see example)
• Multiple audio outputs
• Supports audio inputs for speech-to-text-to-speech
• Keyboard shortcuts for quick messages
• Voice expressions (Microsoft Azure)
• Download messages to audio files
• Translate messages into any language
• Dictionary to translate abbreviations (ily -> I love you (💖))
• Local and remote servers to help you build your own text-to-speech implementations
• Local websocket server to retrieve messages (useful for streaming elements!)

How to use

By default, press Ctrl + Enter to open Izabela.

The text input will be focused allowing you to type your message and send it. The selected text-to-speech engine will translate your text into audio using the selected voice. The audio is then sent through every audio outputs you have configured.

If you have a virtual audio cable installed, you can configure it as an audio output in Izabela then use it as audio input in any application that can receive audio inputs. If you don't have a virtual audio cable installed, you can install one directly from Izabela in the settings.

You can keep using any non-fullscreen application in the background while Izabela is in the foreground. If you're using Izabela over games, I recommend setting your game to borderless, windowed or borderless windowed mode instead of fullscreen.

You can press Ctrl + Enter (or Esc if the text input is focused) to close the window. Pressing Enter while the text input is focused and no text is present will also close the window.

Support

Credentials for the more advanced text-to-speech engines are not provided by default. If you don't want to bother configuring them for each engine, I provide universal API keys to all "Supporter" members or above on my Ko-fi!

You can see the list of engines that are included with universal credentials in the Available engines section.

Note: The Izabela engine is only available to "Supporter" members or above on Ko-fi.

If you prefer configuring your credentials yourself, you can do so by following the guides on the Wiki page.

Available engines

Here's a list of all the text-to-speech engines that are supported in Izabela:

Engine	Included with Universal credentials	Sample	Credits
Izabela (multiple engines)	Yes	Download	https://github.com/Weilbyte/tiktok-tts
Amazon Polly	Yes	Download	https://aws.amazon.com/polly/
Google Cloud TTS	Yes	Download	https://cloud.google.com/text-to-speech
IBM Watson TTS	Temporarily unavailable		https://www.ibm.com/cloud/watson-text-to-speech
Microsoft Azure TTS	Yes	Download	https://azure.microsoft.com/en-us/products/cognitive-services/text-to-speech/
Say	Included by default	Download	https://github.com/Marak/say.js/
Sam	Included by default	Download	https://github.com/discordier/sam / https://github.com/Imrane03/better-sam
Animalese	Included by default	Download	https://github.com/Acedio/animalese.js
ElevenLabs	No	Download	https://beta.elevenlabs.io/speech-synthesis

Guides

Find guides and API documentation on the Wiki page.

Resources

• Figma

Development

The project requires the following in order to run:

• Node 16
• Yarn
• Python 3.10
• Visual Studio Build Tools 2019 or higher
  • Desktop development with C++
  • Node.js build tools