Published On: Mon, Aug 14th, 2017

Google updates the Cloud Speech API with support for some-more languages, word-level timestamps


Google’s Cloud Speech API, that has authorised developers to use Google’s services to register oral difference into content given a launch in 2016, is removing a vital refurbish today.

The many engaging of these is substantially a further of support for 30 new languages on tip of a 89 languages a use already accepted (though, to be fair, Google includes mixed informal variants of English, Spanish and Arabic in a sum count). These new languages embody a likes of Bengali, Latvian and Swahili and, according to Google, cover about a billion speakers.

On tip of this, Google also introduced a few new core facilities to a service. Among these is support for word-level timestamps. The thought here is to tab each word with a timestamp so that developers can, for example, simply concede their users to hear who a given word sounded like. That’s generally engaging for human-augmented transcription and interpretation services that use this API to speed adult their workflows. “Having a ability to map a audio to a content with timestamps significantly reduces a time spent proofreading transcripts,” says Happy Scribe co-founder, André Bastie, whose association uses a use for a $0.10/minute talk transcription service

The kind of files that developers upload to a use can now also be adult to 3 hours prolonged — adult from 80 mins in a prior version. Developers can also ask for a share prolongation to upload files that are even longer.

Like before, developers can get 60 mins of giveaway audio estimate by a Speech API and each additional 15 seconds is billed during $0.006.

About the Author

Leave a comment

XHTML: You can use these html tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>