How to Avoid Robotic Voice Text to Speech Synthesis

Nowadays, more and more people use text-to-speech software to improve their reading efficiency and save time. As an important component of the text-to-speech technology, voices, i.e. speech engines, are the soul of text-to-speech software. Because text-to-speech software must invoke voices to synthetic speech.

However, the default voices shipped with operating systems, for example, Microsoft Sam on Windows XP, are not very natural-sounding. Actually, you need to install more high-quality voices. Fortunately, there are many voices available on the Internet today, such as Microsoft additional voices, Lernout & Hauspie TruVoice and TTS3000 voices, AT&T Natural Voices, Cepstral voices, IVONA voices, etc.. Most of these voices are better than the default voice on your system.

Here is a list of text-to-speech voices available on the Internet.

Microsoft Mike, Mary -

Microsoft Mike and Microsoft Mary are optional male and female voices respectively with better quality, available for download from the Microsoft website or other third party text-to-speech related websites.

The most noticeable feature of these three voices is there are both SAPI 4 and SAPI 5 versions of them. SAPI 5 voices are only available on Windows 2000 and later Windows NT-based operating systems. As an exception, the SAPI 5 versions of these voices cannot be installed on Windows Vista and Windows 7. The SAPI 4 versions are more compatible. They can be installed and used on Windows 98 and all the later Windows operating systems.

Generally, the higher SAPI version, the better quality.

Microsoft Anna -

Microsoft Anna is the default English voice shipped with Windows Vista and Windows 7. Microsoft Anna is a SAPI 5-only voice and is designed to sound more natural than the previous system default voice Microsoft Sam.

This excellent female voice is base on the new Microsoft SAPI 5.3/5.4. So you cannot use it on Windows XP directly and also there is no standalone installer of this voice on the Microsoft website. The only way to use this fantastic voice on Windows XP is by the Microsoft Streets & Trips. The Microsoft Streets & Trips 2006 and later versions install the voice and the new Microsoft SAPI onto Windows XP computers for the voice-prompt direction feature.

Some third party text-to-speech related websites supply smaller repacked Microsoft Anna installers for Windows XP users. However, it's incomplete and not working correctly on Windows XP because the SAPI version of Windows XP is 5.1.

Cepstral Voices -

These high quality voices take up less disk space (average less than 50mb) than most premium voices, do not use as much processor power, and are very fast when creating audio files.

AT&T Natural Voices -

An award winning text to speech technology developed by AT&T Laboratories. Natural Voices is believed by many to be the most natural sounding TTS available, the highly scalable multi-threaded technology can automate and add tremendous interactive and humanistic value to a wide variety of applications and projects.

IVONA Voices -

Human voices for your PC - Use the most natural Text-To-Speech voices ever heard for fun and work.

CereProc Voices -

High-quality voices for computers running Microsoft Windows XP and Vista/7. It can replace the default American English Microsoft voices with a wide range of other accents and languages. CereProc's Windows voices are compatible with Microsoft SAPI5, and all voices are 22khz for excellent clarity and naturalness.

Lernout & Hauspie Voices -

Lernout & Hauspie Speech Products, or L&H, was a leading Belgium-based speech recognition technology company. This company released dozens of high-quality SAPI 4 voices across multiple languages, including ten American English voices and two British English voices.

In addition, Lernout & Hauspie Michael and Michelle are also optional male and female voices licensed by Microsoft from Lernout & Hauspie, and available through Microsoft Office XP and Microsoft Office 2003 or Microsoft Reader.

Most of the above voices can be found on the 2nd Speech Center web site. Click here to download them immediately!