Eric Forbes

I was looking for a way to turn a QuickTime video I just recorded into a text transcript, similar to the video log seen in the movie, Avatar. I had heard about Dragon speech recognition software but discovered it is only available for the PC. There is something for the Mac – MacSpeech Dictate – but I’m not sure if it can generate text from a video source.

I got distracted and discovered a book that describes how to Present like Steve Jobs. Then I came back to my search for the perfect video to text application.

A few weeks ago, I learned about Google’s iPhone Search app that uses their computing cloud to turn audio samples into text in seconds. So I figured, maybe Google would know how to do this.

Now that Google owns YouTube, they’ve started to apply automatic speech recognition (ASR) to many of the videos found on YouTube. YouTube users can upload a text file of the transcript to an existing video they’ve already uploaded and have Google sync the transcript to the video. It can translate into other languages if you want. But how can I turn my own personal video on my MacBook into text? I need to see if they have a solution for that.

Google Accessibility might have the answer. [Not exactly.]

Google labs has a project called Google Audio Indexing (or Gaudi) that uses their speech recognition application to recognize speech contained on YouTube videos. Their first experiment is limited to election videos, but I can see how this might be extended to other videos at some point.

Now how do I access Google’s cloud (or some other application that lives on my Mac) to render my own video to text without uploading my personal video logs YouTube and waiting for them to extend their Gaudi service to the rest of the YouTubiverse? I’ll save that for another day. It’s time to walk the dog…