RoboBlather is a text-to-speech application for Windows.
RoboBlather uses the built-in speech synthesis power of the .NET Framework 3.0 (and newer) to generate speech from whatever text you paste into the edit window. You can control the speed of the voice, render the output to a .wav file, and select any of the installed voices on your machine.
On my development machines, these are the voices that were available:
Windows 10 with all voices installed:
The list of available voices will vary based on the software you have installed on your system, but with the .NET framework installed you should have one or more voices and in many cases new voices can be added.
If you're on Windows 10, you can add the UK English voice Hazel by following these instructions: http://superuser.com/questions/956810/how-do-you-add-the-microsoft-hazel-text-to-speech-voice-back-to-windows-10. In addition, Microsoft Mike, Sam, and Mary can be added using the Microsoft TTS Voices v1.3 installer found in this forum thread (there's also an installer for Anna): http://www.eightforums.com/customization/23916-microsoft-anna-windows-8-a-6.html.
This was originally released at http://zetacentauri.com/software_roboblather.htm and was downloaded more than 25,000 times before being open-sourced.
A prebuilt installer is available in the installer directory here: https://github.com/Xangis/RoboBlather/blob/master/Installer/RoboBlather1.21Setup.exe
- Wave file is recorded as 22050 Hz mono.
This application was written using C# .NET 3.0 and should build on any system that supports .NET 3.0 or newer fully. It has been tested on Windows XP, Vista, 7, 8, and 10. A project is included for Visual Studio 2010, but it should open just fine in newer versions. It doesn't have any dependencies other than the .NET framework.
An installer script for InnoSetup is included in the installer directory, but you'll probably have to modify the paths in order for it to work on your system.
Changes in Version 1.21 (July 2017):
- Added "Load Text" button to load a text file.
Changes in Version 1.2:
- Interface changed from black and yellow to non-inverse (plain) colors.
- Word wrap added to text input box.
Changes in Version 1.1:
- Added window icon.
- Text window accepts multi-line text output.
- Added browse button for output filename.
- All now gives a notification when audio file rendering is complete.
- Text box resizes to grow/shrink with window size.
When rendering speech to a .wav file, you will probably have to close RoboBlather before you can play the generated file. It seems to be one of the quirks of the speech synthesizer that it doesn't close a file when audio is rendered (because it expects to keep saying things). This may be fixable via a code change, but I haven't really looked into it.
There was some effort made to use the Desktop App Converter to package this for the Windows Store, but it never quite made it. There's a package.bat file in the installer directory with progress so far.
I consider this program "done" and feature-complete, but do still maintain it and will accept pull requests if you have an improvement to contribute.
I do sometimes still use it to generate .wav files of robot speech. To my knowledge, it's the quickest, easiet, and free-est way to do so on a Windows machine. That means that it's very likely that I'll update it if a new Windows version makes it necessary.