Tag Archives: machine learning

Using A.I. to Create Music with Ampermusic and Jukedeck

For the last 14 years I’ve created the Audio Art Tour for Burning Man. It’s kind of a docent led audio guide to the major art installations out there, similar to an audio guide you might get at a museum.

Burning Man always has a different ‘theme’ and this year it was ‘I, Robot’. I generally try and find background music related to the theme. EDM is big at Burning Man, land of 10,000 DJs, so I could’ve just grabbed some electronic tracks that sounded robotic. Easy enough to do. However I  decided to let Artificial Intelligence algorithms create the music! (You can listen to the tour and hear the different tracks)

This turned out to be not so easy, so I’ll break down what I had to do to get seven unique sounding, usable tracks. I had a bit more success with AmperMusic, which is also currently free (unlike Jukedeck), so I’ll discuss that first.

Getting the Tracks

The problem with both services was getting unique sound tracks. The A.I. has a tendency of creating very similar sounding music. Even if you select different styles and instruments you often end up with oddly similar music. This problem is compounded by Amper’s inability to render more than about 30 seconds of music.

Using Artificial Intelligence and machine learning to create music

What I found I had to do was let it generate 30 seconds randomly or with me selecting the instruments. I did this repeatedly until I got a 30 second sample I liked. At which point I extended it out to about 3 or 4 minutes and turned off all the instruments but two or three. Amper was usually able to render that out. Then I’d turn off those instruments and turn back on another three. Then render that. Rinse, repeat until you’ve rendered all the instruments.

Now you’ve got a bunch of individual tracks that you can combine to get your final music track. Combine them in Audition or even Premiere Pro (or FCP or whatever NLE) and you’re good to go. I used that technique to get five of the tracks.

Jukedeck didn’t have the rendering problem but it REALLY suffered from the ‘sameness’ problem. It was tough getting something that really sounded unique. However, I did get a couple good tracks out of it.

Problems Using Artificial Intelligence

This is another example of A.I. and Machine Learning that works… sort of. I could have found seven stock music tracks that I like much faster (this is what I usually do for the Audio Art Tour).  The amount of time it took me messing around with these services was significant. Also, if Jukedeck is any indication, a music track from one of these services will cost as much as a stock music track. Just go to Pond5 to see what you can get for the same price. With a much, much wider variety. I don’t think living, breathing musicians have much to worry about. At least for now.

That said, I did manage to get seven unique, cool sounding tracks out of them. It took some work, but it did happen.

As with most A.I./ML, it’s difficult to see what the future looks like. There has certainly been a ton of advances, but I think in a lot of cases, it’s some of the low hanging fruit. We’re seeing that with Speech-to-text algorithms in Transcriptive where they’re starting to plateau and cluster around the same accuracy levels. The fruit (accuracy) is now pretty high up and improvement are tough. It’ll be interesting to see what it takes to break through that. More data? Faster servers? A new approach?

I think music may be similar. It seems like it’s a natural thing for A.I. but it’s deceptively difficult to do in a way that mimics the range and diversity of styles and sounds that many human musicians have. Particularly a human armed with a synth that can reproduce an entire orchestra. We’ll see what it takes to get A.I. music out of the Valley of Sameness.

 

Artificial Intelligence vs. Video Editors

With Transcriptive, our new tool for doing automated transcriptions, we’ve dove into the world of A.I. headfirst. So I’m pretty familiar with where the state of industry is right now. We’ve been neck deep in it for the last year.

A.I. is definitely changing how editors get transcripts and search video for content. Transcriptive demonstrates that pretty clearly with text.  Searching via object recognition is something that also is already happening. But what about actual video editing?

One of the problems A.I. has is finishing. Going the last 10% if you will. For example, speech-to-text engines, at best, have an accuracy rate of about 95% or so. This is about on par with the average human transcriptionist. For general purpose recordings, human transcriptionists SHOULD be worried.

But for video editing, there are some differences, which are good news. First, and most importantly, errors tend to be cumulative. So if a computer is going to edit a video, at the very least, it needs to do the transcription and it needs to recognize the imagery. (we’ll ignore other considerations like style, emotion, story for the moment) Speech recognition is at best 95%, object recognition is worse. The more layers of AI you have, usually those errors will multiply (in some cases there might be improvement though) . While it’s possible automation will be able to produce a decent rough cut, these errors make it difficult to see automation replacing most of the types of videos that pro editors are typically employed for.

Secondly, if the videos are being done for humans, frequently the humans don’t know what they want. Or at least they’re not going to be able to communicate it in such a way that a computer will understand and be able to make changes. If you’ve used Alexa or Echo, you can see how well A.I. understands humans. Lots of situations, especially literal ones (find me the best restaurant), it works fine, lots of other situations, not so much.

Many times as an editor, the direction you get from clients is subtle or you have to read between the lines and figure out what they want. It’s going to be difficult to get A.I.s to take the way humans usually describe what they want, figure out what they actually want and make those changes.

Third… then you get into the whole issue of emotion and storytelling, which I don’t think A.I. will do well anytime soon. The Economist recently had an amusing article where it let an A.I. write the article. The result is here. Very good at mimicking the style of the Economist but when it comes to putting together a coherent narrative… ouch.

It’s Not All Good News

There are already phone apps that do basic automatic editing. These are more for consumers that want something quick and dirty. For most of the type of stuff professional editors get paid for, it’s unlikely what I’ve seen from the apps will replace humans any time soon. Although, I can see how the tech could be used to create rough cuts and the like.

Also, for some types of videos, wedding or music videos perhaps, you can make a pretty solid case that A.I. will be able to put something together soon that looks reasonably professional.

You need training material for neural networks to learn how to edit videos. Thanks to YouTube, Vimeo and the like, there is an abundance of training material. Do a search for ‘wedding video’ on YouTube. You get 52,000,000 results. 2.3 million people get married in the US every year. Most of the videos from those weddings are online. I don’t think finding a few hundred thousand of those that were done by a professional will be difficult. It’s probably trivial actually.

Same with music videos. There IS enough training material for the A.I.s to learn how to do generic editing for many types of videos.

For people that want to pay $49.95 to get their wedding video edited, that option will be there. Probably within a couple years. Have your guests shoot video, upload it and you’re off and running. You’ll get what you pay for, but for some people it’ll be acceptable. Remember, A.I. is very good at mimicking. So the end result will be a very cookie cutter wedding video. However, since many wedding videos are pretty cookie cutter anyways… at the low end of the market, an A.I. edited video may be all ‘Bridezilla on A Budget’ needs. And besides, who watches these things anyways?

Let The A.I Do The Grunt Work, Not The Editing

The losers in the short term may be assistant editors. Many of the tasks A.I. is good for… transcribing, searching for footage, etc.. is now typically given to assistants. However, it may simply change the types of tasks assistant editors are given. There’s a LOT of metadata that needs to be entered and wrangled.

While A.I. is already showing up in many aspects of video production, it feels like having it actually do the editing is quite a ways off.  I can see creating A.I. tools that help with editing: Rough cut creation, recommending color corrections or B roll selection, suggesting changes to timing, etc. But there’ll still need to be a person doing the edit.

 

How Doc Filmmakers Are using A.I. to Create Captions and Search Footage in Premiere Pro

Artificial Intelligence (A.I.) and machine learning are changing how video editors deal with some common problems. 1) how do you get accurate transcriptions for captions or subtitles? And 2) how do you find something in hours of footage if you don’t know exactly where it is?

Getting out of the Transcription Dungeon

Kelley Slagle, director, producer and editor for Cavegirl Productions, has been working on Eye of the Beholder, a documentary on the artists that created the illustrations for the Dungeons and Dragon game. With over 40 hours of interview footage to comb through searching through it all has been made much easier by Transcriptive, a new A.I. plugin for Adobe Premiere Pro.


eye-beholder 

Why Transcribe?

Imagine having Google for your video project. Turning all the dialog into text makes everything easily searchable (and it supports 28 languages). Not too mention making it easy to create captions and subtitles.

The Dragon of Time And Money

Using a traditional transcription service for 40 hours of footage, you’re looking at a minimum of $2400 and a few days to turn it all around. Not exactly cost or time effective. Especially if you’re on a doc budget. However, it’s a problem for all of us.

Transcriptive helps solve the transcription problem, and the problems of searching video and captions/subtitles. It uses A.I. and machine learning to automatically generate transcripts with up to 95% accuracy and bring them into Premiere Pro. And the cost? About $4/hour (or much less depending on the options you choose) So, 40 hours is $160 vs $2400. And you’ll get all of it back in a few hours.

Yeah, it’s hard to believe.

Read what these three filmmakers have to say and try the Transcriptive demo out on your own footage. It’ll make it much easier to believe.

 

“We are using Transcriptive to transcribe all of our interviews for EYE OF THE BEHOLDER. The idea of paying a premium for that much manual transcription was daunting. I am in the editing phase now and we are collaborating with a co-producer in New York. We need to share our ideas for edits and content with him, so he is reviewing transcripts generated by Transcriptive and sending us his feedback and vice versa. The ability to get a mostly accurate transcription is fine for us, as we did not expect the engine to know proper names of characters and places in Dungeons & Dragons.” – Kelley Slagle, Cavegirl Productions

Google Your Video Clips and Premiere Project?

 

Since everything lives right within Premiere, all the dialog is fully searchable. It’s basically a word processor designed for transcripts, where every word has time code. Yep, every word of dialog has time code. Click on the word and jump to that point on the timeline. This means you don’t have to scrub through footage to find something. Search and jump right to it. It’s an amazing way for an editor to find any quote or quip.

As Kelley says, “We are able to find what we need by searching the text or searching the metadata thanks to the feature of saving the markers in our timelines. As an editor, I am now able to find an exact quote that one of my co-producers refers to, or find something by subject matter, and this speeds up the editing process greatly.”

Joy E. Reed of Oh My! Productions, who’s directing the documentary, ‘Ren and Luca’ adds, “We use sequence markers to mark up our interviews, so when we’re searching for specific words/phrases, we can find them and access them nearly instantly. Our workflow is much smoother once we’ve incorporated the Transcriptive markers into our project. We now keep the Markers window open and can hop to our desired areas without having to flip back and forth between our transcript in a text document and Premiere.”

Workflow, Captions, and Subtitles

ren-luca-L

Captions and subtitles are one of the key uses of Transcriptive. You can use it with the Premiere’s captioning tool or export many different file formats (SRT, SMPTE, SCC, MCC, VTT, etc) for use in any captioning application.

“We’re using Transcriptive to transcribe both sit down and on-the-fly interviews with our subjects. We also use it to get transcripts of finished projects to create closed captions/subtitles.”, says Joy. “We can’t even begin to say how useful it has been on Ren and Luca and how much time it saves us. The turnaround time to receive the transcripts is SO much faster than when we sent it out to a service. We’ve had the best luck with Speechmatics. The transcripts are only as accurate as our speakers – we have a teenage boy who tends to mumble, and his stuff has needed more tweaking than some of our other subjects, but it has been great for very clearly recorded material. The time it saves vs the time you need to tweak for errors is significant.”

captions

Transcriptive is fully integrated into Premiere Pro, you never have to leave the application or pass metadata and files around. This makes creating captions much easier, allowing you to easily edit each line while playing back the footage. There are also tools and keyboard shortcuts to make the editing much faster than a normal text editor. You then export everything to Premiere’s caption tool and use that to put on the finishing touches and deliver them with your media.

Another company doing documentary work is Windy Films. They are focused on telling stories of social impact and innovation, and like most doc makers are usually on tight budgets and deadlines. Transcriptive has been critical in helping them tell real stories with real people (with lots of real dialog that needs transcribing).

They recently completed a project for Planned Parenthood. The deadline was incredibly tight. Harvey Burrell, filmmaker at Windy, says, “We were trying to beat the senate vote on the healthcare repeal bill. We were editing while driving back from Iowa to Boston. The fact that we could get transcripts back in a matter of hours instead of a matter of days allowed us to get it done on time. We use Transcriptive for everything. The integration into premiere has been incredible. We’ve been getting transcripts done for a long time. The workflow was always a clunky; particularly to have transcripts in a word document off to one side. Having the ability to click on a word and just have Transcriptive take you there in the timeline is one of our favorite features.”

Getting Accurate Transcripts using A.I.

 

Audio quality matters. So the better the recording and the more the talent enunciates correctly, the better the transcript. You can get excellent results, around 95% accuracy, with very well recorded audio. That means your talent is well mic’d, there’s not a lot of background noise and they speak clearly. Even if you don’t have that, you’ll still usually get very good results as long as the talent is mic’d. Even accents are ok as long as they speak clearly. Talent that’s off mic or if there’s crosstalk will cause it to be less accurate.

6-Full-Screen

Transcriptive lets you sign up with the speech services directly, allowing you to get the best pricing. Most transcription products hide the service they’re using (they’re all using one of the big A.I. services), marking up the cost per minute to as much as .50/min. When you sign up directly, you get Speechmatics for $0.07/min. And Watson gives you the first 1000 minutes free. (Speechmatics is much more accurate but Watson can be useful).

Transcriptive itself costs $299 when you check out of the Digital Anarchy store. A web version is coming soon as well. To try transcribing with Transcriptive you can download the trial version here. (remember, Speechmatics is the more accurate service and the only service available in the demo) Reach out to sales@nulldigitalanarchy.com if you have questions or want an extended trial.

Transcriptive is a plugin that many didn’t know they were waiting for. It is changing the workflow of many editors in the industry. See for yourself how we’re transforming the art of transcription.