In a move meant to boost accessibility and search engine relevance, Google has announced an automatic captions service for YouTube videos, using the "automatic speech recognition" (ASR) technology currently used in the transcription feature of Google Voice.
Although YouTube has long offered options for content providers to upload their own captions and subtitles, few have chosen to do so, with just around 100,000 captioned videos available out of hundreds of millions of clips posted on the site.
Although Google admits that the machine-generated captions "will not always be perfect," they see the introduction of the technology as a positive step in providing content to deaf and hearing disabled users. Users of Google Voice have long had mixed results with ASR, even posting the most humorous transcription errors online for others to see. Aware of such "amusing" examples, the Google team is confident that their voice recognition algorithms "will continue to improve with time."
Also announced to coincide with the automatic transcription service is a syncing feature for those who have manually created their own captions. Called "automatic caption timing," the system will allow YouTube users to upload simple text files containing their subtitles that can then be adjusted to fit the pacing of the video online.
The effectiveness of these services, and their reception among the YouTube community, will become more apparent when they are made available "at the end of this week."