MagicTTS ™ Overview

MagicTTS™ offers state of the art speech technologies for generating customized voices for desktop applications. It combines manipulations of spoken utterances created with Text-To-Speech engines and of utterances in popular wave file format in a single software application, enabled with easy-to-use voice parameter control tool set. You can build a single synthetic voice using with TTS engines provided with MagicTTS™, and then modify in near real-time various voice aspects including the speaking speed, pitch intonation, and voice timber, to create multiple voices, each with different speaking style. Currently, Festival® TTS or HILTTS® engines are provided as sample engines. You can also choose to modify your proprietary voices using the same tool set, making MagicTTS™ a fast, versatile, and convenient solution for speech production.

What can MagicTTS™ do?

  • State-of-the-art voice adaptation
  • Whether using your proprietary speech audio file or TTS generated speech, MagicTTS™ can modify voice styles to generate new voices while preserving the high quality of the incoming speech.

  • Provides one-stop solution for speech production.
  • You can select a TTS engine, select a voice, type texts to be spoken, and then fine-tune your speech output, all using GUI-based controls in simple steps.

  • Flexible toolset for on-going speech productions
  • You have several options of saving your TTS engine-based speech output; You can save them as an audio file in .raw or .wav file format; you can store the TTS voice engine and voice parameter settings in a data file for a future reproduction; you can add the settings to the GUI's 'Preset Voices' list.

  • Interfaces with existing TTS systems
  • Any commercial TTS systems such as AT&T, Bell Labs, or Loquendo's TTS can be incorporated.