{"id":5235,"date":"2017-04-06T14:04:38","date_gmt":"2017-04-06T21:04:38","guid":{"rendered":"https:\/\/blogs.msdn.microsoft.com\/translation\/?p=5235"},"modified":"2017-04-06T14:04:38","modified_gmt":"2017-04-06T21:04:38","slug":"japanese-becomes-the-10th-speech-translation-language-supported-by-microsoft-translator","status":"publish","type":"post","link":"https://www.microsoft.com\/en-us\/translator/blog\/2017\/04\/06\/japanese-becomes-the-10th-speech-translation-language-supported-by-microsoft-translator\/","title":{"rendered":"Japanese becomes the 10th speech translation language supported by Microsoft Translator"},"content":{"rendered":"

\"Tokyo<\/a><\/p>\n

Today, Microsoft Translator announces the availability of its 10th<\/sup> speech translation language: Japanese. This new language is now available across all Microsoft Translator supported technologies and products along with the already released nine other speech translation languages<\/a>: Arabic, Chinese, English, French, German, Italian, Portuguese, Russian and Spanish.<\/p>\n

Microsoft Translator is the first end-to-end speech translation solution optimized for real-life conversations (vs. simple human to machine commands) available on the market. It is one of the services in Microsoft\u2019s portfolio of artificial intelligence technologies<\/a> designed to make AI accessible to all.<\/p>\n

Speech translation is a hard problem to solve, as is always the case when machines are trying to mimic a deeply human capability.\u00a0 It uses two neural-network based Artificial Intelligence (AI) technologies: Automatic Speech Recognition and Machine Translation. It also uses a unique natural language processing technology (TrueText) and a speech synthesizer, aka \u201cText to Speech\u201d, which enables users to hear, and not only read, the translation.<\/p>\n

These technologies are then connected to perform the speech translation function:<\/p>\n

\"english_to_japanese_1920\"<\/a><\/p>\n

 <\/p>\n

    \n
  1. The sound is transcribed into text through the speech recognition AI<\/li>\n
  2. TrueText then processes this text to remove unnecessary speech elements, such as redundant words or fillers like \u201cum\u201d (English) or \u201ceto\u201d (Japanese), that would generate poor translations<\/li>\n
  3. The machine translation AI<\/a> then translates each word using the context of the full sentence<\/li>\n
  4. Finally, the text to speech generates the audio output from this translated text<\/li>\n<\/ol>\n

     <\/p>\n

    With this release, businesses, developers, and end users alike will be able to use Japanese in the various apps and services offered or powered by Microsoft Translator:<\/p>\n