As a long-time advocate for people with hearing loss, I commend Google and YouTube for making strides in the area of online video accessibility. As the owner of a realtime transcription company (and a realtime captioner myself) I can tell you that the machine-generated automatic captions will not be good enough to provide meaningful accessibility for many years to come. At best, the results are approx. 60-65% accurate. Captioning standards require at least 98% accuracy. I’ve been watching this technology closely, since I am eager to use this automatic speech-to-text technology in certain aspects of my transcription business. The reality is, it’s nowhere near ready.

We have developed a new media player that enables searching of video based on transcripts, and we do have captioning features on the player. I am very concerned about providing the BEST access possible for those folks with hearing loss. In my opinion, videos transcribed and captioned by trained humans are still necessary for full accessibility.

Is “something better than nothing” when it comes to captioning? Yes, when we’re speaking of videos being produced and posted by individuals. But when we’re speaking of content that is produced by businesses and universities and governmental entities, (some of whom must comply with Section 508 of the Rehabilitation Act that requires accessibility on their websites), the automatic machine-generated captions are not good enough. After all, the objective is to provide EQUAL access.

If you’d like to check out our searchable online media player, please take a look at our featured video about a deaf man in South Florida.
http://www….

The primary focus of our player is SEARCHABILITY of video, from within the player and from the internet. We are working to make the captioning features better, and we welcome your comments and feedback (on the player and on our transcription accuracy).

Thank you for writing about this topic.

Tanya English, Dir. of Technology
Realtime Transcription, Inc.

http://www…