編集

次の方法で共有


Speech settings

Languages are spoken in different ways around the world, even among the speakers of one language. Let’s consider Arabic as an example. There are global conventions for unifying the writing of the language, such as the Modern Standard Arabic (MSA), but there are also several spoken varieties of Arabic, usually at a regional level. This is also true for Spanish, Portuguese, and many others. Language varieties, or dialects, can be regional or social. They differ by pronunciation, vocabulary, or even grammar.

Speech settings for operating systems include voice recognition and text-to-speech (TTS). The language chosen for voice recognition is the language that the users want to speak with their devices. Voice recognition engines are usually tailored for spoken varieties of a given language. Operating systems can also have an additional setting that allows the system to recognize non-native accents for the selected language.

If operating systems or applications provide TTS, they can provide users with a variety of voices to choose from, to match their preferred regional variant or the speaker’s gender. The operating systems or applications might also support installing additional voice packages if the users cannot find what they want among the default voices.