Google Text-to-Speech is a screen reader application developed by Google for its Android operating system. It powers applications to read aloud (speak) the text on the screen which support many languages. Text-to-Speech may be used by apps such as Google Play Books for reading books aloud, by Google Translate for reading aloud translations providing useful insight to the pronunciation of words, by Google Talkback and other spoken feedback accessibility-based applications, as well as by third-party apps. Users must install voice data for each language.
Video Google Text-to-Speech
Supported languages
Currently (version x.x), languages supported by Google Text-to-Speech include, Bengla (Bangladesh), Bengla (India), Cantonese (Hong Kong), Czech, Danish, Dutch, English (Australia), English (India), English (United Kingdom), English (United States), Estonian, Filipino, Finnish, French (France), French (Canada), German, Greek, Hindi, Hungarian, Indonesian, Italian, Japanese, Javanese, Khmer, Korean, Mandarin (China), Mandarin (Taiwan), Nepali, Norwegian, Polish, Portuguese (Brazil), Romanian, Russian, Sinhala, Slovak, Spanish (Spain), Spanish (United States), Swedish, Thai, Turkish, Ukrainian and Vietnamese.
Maps Google Text-to-Speech
Evolution
Some app developers have started adapting and tweaking their Android Auto apps to include Text to Speech, such as Hyundai in 2015. Apps such as textPlus and WhatsApp use Text to Speech to read notifications aloud and provide voice-reply functionality. Cloud Text-to-Speech is powered by WaveNet, software created by Google's UK-based AI subsidiary DeepMind. This is significant for two reasons. First, ever since Google bought DeepMind in 2014, it's been exploring ways to turn the company's AI talent into tangible products. So far, this has meant using DeepMind's algorithms to reduce electricity costs for cooling in Google's data centers by 40 percent and DeepMind's forays into health care. But, directly integrating WaveNet into its cloud service is arguably more significant, especially as Google tries to win cloud business away from Amazon and Microsoft, presenting its AI skills as its differentiating factor. Second, DeepMind's AI voice synthesis tech is some of the most advanced and realistic in the business. Most voice synthesizers (including Apple's Siri) use what's called concatenative synthesis, in which a program stores individual syllables -- sounds such as "ba," "sht," and "oo" -- and pieces them together on the fly to form words and sentences. This method has gotten pretty good over the years, but it still sounds stilted. WaveNet, by comparison, uses machine learning to generate audio from scratch. It actually analyzes the waveforms from a huge database of human speech and re-creates them at a rate of 24,000 samples per second. The end result includes voices with subtleties like lip smacks and accents. When Google first unveiled WaveNet in 2016, it was far too computationally intensive to work outside of research environments, but it's since been slimmed down significantly, showing a clear pipeline from research to product.
Version history
November 2013
- Korean now supported.
March 2014
- Google announced that Arabic language will never be supported despite having more than 467 million native spekers.
- Version 3.0 added support for natural high-quality voices.High quality voices now featured in English (United States) as Female (high quality) whilst English (United Kingdom) also now featured three new high quality voices; Male, Female (high quality) and Male (high quality). These new high quality voices are much larger than the prior versions in terms of file size with 244MB for English US female (high quality) compared to just 6.8MB for the regular female voice version. These high quality voices were added to ensure higher quality pronunciation and enunciation with intonations that are more natural.
- Support for Brazilian, Portuguese and Spanish (United States) bringing the total number of languages supported to nine at this point. (German, English (UK), English (US), Spanish (ES), Spanish (US), French, Italian, Korean, and Portuguese (BR). Only English (US) and English (UK) have high-quality voice packs for now.) German, English UK, English US, Spanish ES, Spanish US, French, Italian, Korean, and Portuguese (BR). Only English US and English UK have high-quality voice packs for now.
- User Interface tweaks: Due to having multiple voices for some languages a toggle was added to voices with 2 or more voice packs.
May 2014
- Russian, Dutch, Polish and English (Indian) added to the currently supported list of languages.
September 2014
- Support for Japanese output added.
December 2014
- Version 4 Available (For 6.0 Marshmallow and up)
- Support for Hindi and Indonesian output.
- Improved output quality. Standard quality voices now surpass the quality of the high quality voices from previous releases.
July 2015
- Four new languages now supported: Cantonese (Hong Kong), Mandarin (China), Thai (Thailand) and Turkish (Turkey)
- Bug fixes and other improvements.
February 2016
- Improved voice quality
- Added support for Bengali (Bangladesh), Danish (Denmark), English (Australia), Finnish (Finland), Hungarian (Hungary), Norwegian (Norway), and Mandarin (Taiwan) and Swedish.
- The offline voices can now speak at a faster rate.
- Plus lots of bug fixes and performance improvements.
June 2016
- Added support for Swedish and Vietnamese.
- Bug fixes and improvements
October 2016
- Alternative voice variations now available on every device.
- Added support to amplify speech volume over other audio.
- Extended support for emoji verbalisation in Chinese, Dutch, Danish, English, French, German, Italian, Japanese, Korean, Polish, Portuguese, Russian and Spanish.
- Bug fixes and improvements.
April 2017
- Added support for Bengali (India), Czech, Khmer, Nepali, Sinhala and Ukrainian.
- Number processing can now be turned off in settings. This produces a more literal pronunciation of the text. For example 09/10/2017 will be pronounced as oh nine slash ten... Only available for English voices.
- Intonation control is now available for more voices.
- Various other improvements to various voices.
October 2017
- Added support for Filipino and Greek.
January 2018
- Added support for Estonian, Romanian and Slovak.
- Various other improvements to our voices.
March 2018
- Added support for Estonian, Romanian and Slovak.
- Various other improvements to our voices.
May 2018
- Added support for Estonian, Romanian and Slovak.
- Various other improvements to our voices.
July 2018
- Added support for French (Canada), Sundanese.
- Various other improvements to our voices.
See also
- Speech synthesis
- VoiceOver
References
Source of the article : Wikipedia
0 Comments