 |
 |
I haven't tried it yet, but the Open AI foundation recently released an open source neural net called Whisper for transcribing/translating English audio. Here's the page: https://openai.com/blog/whisper/ And here's an article about it: https://arstechnica.com/information-technology/2022/09/new-ai-model-from-openai-automatically-recognizes-speech-and-translates-to-english/ I haven't experimented with it yet, but I'm kind of interested to try it. Will -----Original Message----- From: Code for Libraries <[log in to unmask]> On Behalf Of Dan Johnson Sent: Friday, October 21, 2022 1:21 PM To: [log in to unmask] Subject: Re: [CODE4LIB] video to text Thank you, Peter Murray, for the fascinating AWS Transcribe writeup. In case someone is interested in going down that route, I did have success, some years ago, taking a JSON file someone else had generated from AWS Transcribe, and converting it into a very readable .docx. It requires only a simple two-line Python script with the tscribe library. Information here: https://github.com/kibaffo33/aws_transcribe_to_docx On Fri, Oct 21, 2022 at 2:10 PM Peter Murray < [log in to unmask]> wrote: > I did something like this last month for creating transcripts from > podcasts using Amazon Transcribe. Details and links to the code here: > https://dltj.org/article/generating-podcast-transcripts/ > > > Peter > > From: Dan Johnson <[log in to unmask]> <[log in to unmask]> > Reply: Code for Libraries <[log in to unmask]> > <[log in to unmask]> > Date: October 21, 2022 at 2:01:30 PM > To: [log in to unmask] <[log in to unmask]> > <[log in to unmask]> > Subject: Re: [CODE4LIB] video to text > > If your university gives you an Office 365 account (and Notre Dame > does), Word 365 will transcribe up to 300 minutes of audio per month > from a sound file in the .wav, .mp4, .mpa, or .mp3 formats. In my own > (admittedly minor) tinkering, I've been surprised at how good the > transcript is. Microsoft has a 90 second tutorial here: < > > https://support.microsoft.com/en-us/office/transcribe-your-recordings- > 7fc2efec-245e-45f0-b053-2a97531ecf57 > >. > > If you're handy with AWS, you can also use Amazon Transcribe ( > https://aws.amazon.com/transcribe/), but that is much more involved. I > have no experience myself, though some colleagues have had success > with larger projects there. > > Best, > Dan > > > On Fri, Oct 21, 2022 at 1:58 PM Lolis, John <[log in to unmask]> > wrote: > > > I don't have technology to offer for that purpose, but if you decide > > to > go > > with a service, I can tell you that I've found Amara to be very > affordable, > > of excellent quality and fantastic customer service. I used them to > > not only caption videos but to also translate them from several > > languages. I couldn't have asked for a better experience with them, > > and that was after some back and forth working things out over the extra languages. > > > > https://amara.org/ > > > > John Lolis > > Coordinator of Computer Systems > > > > 100 Martine Avenue > > White Plains, NY 10601 > > > > tel: 1.914.422.1497 > > fax: 1.914.422.1452 > > > > https://whiteplainslibrary.org/ > > > > *“I would rather have questions that can’t be answered than answers > > that can’t be questioned.”* — Richard Feynman < > > > > https://click.fourhourmail.com/5qure95xkf7hvvo93wh2/7qh7h8h05vr4zrtz/a > HR0cHM6Ly9lbi53aWtpcGVkaWEub3JnL3dpa2kvUmljaGFyZF9GZXlubWFu > > >, > > theoretical physicist and recipient of the Nobel Prize in Physics in > > 1965 > > > > > > On Fri, 21 Oct 2022 at 13:20, Eric Lease Morgan <[log in to unmask]> wrote: > > > > > Do you know of a video to text applications? I colleague asked me: > > > > > > I have four video recordings of conference sessions and wonder if > > > there is a tool or technology that will help me transcribe these > > > into the written word? > > > > > > Do y'all have any suggestions or experience in this regard? > > > > > > -- > > > Eric Morgan > > > University of Notre Dame > > > > > > > > -- > *Daniel Johnson, Ph.D.* > *English; Digital Humanities**; and Film, Television, and Theatre * > *Librarian* > *Navari Family Center for Digital Scholarship, **Hesburgh Libraries* > > *University of Notre Dame* > 250C Hesburgh Library > Notre Dame, IN 46556 > o: 574-631-3457 > e: [log in to unmask] > -- *Daniel Johnson, Ph.D.* *English; Digital Humanities**; and Film, Television, and Theatre * *Librarian* *Navari Family Center for Digital Scholarship, **Hesburgh Libraries* *University of Notre Dame* 250C Hesburgh Library Notre Dame, IN 46556 o: 574-631-3457 e: [log in to unmask]