Jump to content

Search the Community

Showing results for tags 'dialog'.

  • Search By Tags

    Type tags separated by commas.
  • Search By Author

Content Type


  • Main Board
    • Current
    • The Daily Journal
    • General Discussion
    • Equipment
    • Cameras... love them, hate them
    • Recording Direct to Computer
    • Workflow
    • The Post Place
    • Images of Interest
    • Macs... and the other computer
    • All Things Apple
    • Technical Reference
    • Do It Yourself
    • Manufacturers & Dealers
    • Work Available - Available for Work
    • Post to the Host
    • Donate to Support JWSOUNDGROUP

Product Groups

  • Widgets

Find results in...

Find results that contain...

Date Created

  • Start


Last Updated

  • Start


Filter by number of...


  • Start





Website URL








Found 6 results

  1. This mummy isn't a movie, but an actual mummified Egyptian priest from 3000 years ago. Researchers wheeled his remains into a CT scanner, mapped his vocal tract, and 3D-printed a replica of his throat and mouth. They bolted it to a compression horn driver, and claim it reproduces the guy's authentic voice. It makes a good story in today's NYTimes along with a brief audio sample and a link to the research paper in Nature. It's also only a story. The mechanism used to create "speech" -- a complex vowel waveform generated by a computer and controlled by a joystick, then sent through a few fixed low-Q resonators in the 3D-printed "mouth" -- is nothing like the way human voices work. We have high-Q resonators, constantly moving to form different filters on the wideband buzz from the vocal folds. Essentially, they've put a non-linear horn on the output of a conventional speech synthesizer. And it's just guesswork, because we have no idea what sounds (or resonances) were used in the priest's language. The real breakthrough is mapping and printing an ancient mouth, even if it doesn't have the muscles essential for speech. It's a bit more sophisticated than getting a new set of false teeth... but how they made this mouth "talk" is just window dressing. So why am I posting this admittedly speech-nerd story at JWSoundGroup? Aside from the scientific interest (yes, I do get off on this stuff)... Some producer is going to glance at the Times' article, and then demand we create authentic voices for their next horror film or biblical epic!
  2. Fascinating NYT article this morning, https://nyti.ms/2yVqUS8 Researchers put beat-boxers into an MRI, and recorded the motions while they were making full drum-kit sounds vocally. Amazing mouth movements, some of which are like speech... and others, which don't seem to appear in any human language! Lessons for anyone interested in how the human instrument actually works. Which, after all, is what we spend most of our time documenting. ...the researchers apparently were looking for the same kind of information, to help develop language-independent speech recognition and synthesis.
  3. Lyrebird.ai is a Canadian company doing AI creation of new speech from sample recordings. They have an online demo, where you record a minimum of one minute (they guide you through sentences, so the samples have a key), it runs the samples through a neural net, and then it'll create your voice saying anything you type. Pros: almost real-time, with a web interface. Cons: still somewhat artificial sounding, but a lot better than previous while-you-wait examples. ... and this is just a first-gen beta demo, with a really small training set and no ability to tweak the results. The company isn't posting anything about their technique, so I'm just guessing (from the operation and from the principals' bios) that it's NN. They seem to be interested in selling their sdk to other developers, rather than offering a service to filmmakers... but that's also just a guess.
  4. Hello one and all, I recently saw the movie Mud. I found it quite enjoyable on many levels. Good story, great acting - particularly notable kid actors, a soundtrack that perfectly fit the vibe. As a movie it is a great blend of Mark Twain and Stand By Me. I was struck by how good the dialog sounded. And there were many difficult scenes: on small boats with engines running, actors in trees, around lots of water, or any number of situations like that. Best I can tell via IMDB the production sound mixer was Ethan Andrus. Anyone know much about the onset work? How much ADR? Lots of wires or multiple booms? Just curious. Cheers, Sam
  5. So I have a feature that has a fair bit of outdoor shots that were filmed on location at a catholic school on E Colorado with traffic driving by. They won't ADR so I'm going to try and do a little multiband expansion to tame that noisy beast. i also have run rx2 on a few of the noisiest sections to try to at least get the noise floor to match a little betters making other fades between cuts a little more seamless. Now I've never really used waves c4 so I'd appreciate some general tips you guys have for me. Tell me how you like to use this tool?
  6. Do you guys have any favorite Altiverb tweaks you do in the dialog mix? I've been working on this feature and use it quite a big to help get the dialog cuts to blend better as well as giving it a better, more cinematic feel within the room. But I always wonder if there's something I could be doing to help make it even better so I thought I'd ask people smarter than me! Thanks for any advice you can share! John
  • Create New...