This past summer, I took on a fill-in show at our local campus/community radio station. A traditional jazz show, but I decided to only ever say the station's call letters, frequency, and the word "jazz". Over and over and over again. [more inside]
The sample chapter linked here is from Visualizations for Heartbreak, which I self-published in 2020 (in print and electronic versions). The narrative of the book focuses on the traumatic ending of a romantic relationship, but the anguish, rage, shame, and catharsis depicted in its pages are also relevant to other experiences with difficult good-byes. The text is accompanied by photographs and collages (also by the author), and I've recently had recordings made of me reading the text. I believe there is a need for this book, but so far I haven't had much success marketing it. Any suggestions are welcome!
AI Jukebox is a fascinating project from OpenAI that uses cutting-edge neural neworks to perform all sorts of musical magic -- it can take a clip of a song and continue it in a new way, sing text lyrics in any artist's voice, make a song sound like it's being sung by someone else. My favorite? Tell it to generate music by an artist without any other info, and it will produce a gibberish song with nonsense lyrics... that still sounds 100% real and just like the actual singer or band with their unique style. You can hear instruments, melodies, sometimes an audience, the breathing of the lead singer -- but the whole thing is generated completely from scratch by the AI, not with samples or digital sounds. It's not flawless -- some of the songs ramble, with glitchy effects or a mutating voice. But these just add to the vibe, like it's from a dream or a parallel universe. I went through their database to find the best examples of these tracks from the most famous artists, then turned them into an audio quiz on Sporcle -- complete with AI-generated art of the artists I made to serve as hints in the second round. How many of the artists can you name? [more inside]
We made some cool new machine learning models for separating birdsong in soundscape recordings, and demonstrated how to use the separated audio to improve downstream classification. The separation model is available on github, where we've also got lots more examples. There's also a paper.
For the last year, I've worked on Skittish, a weird space for virtual events and gatherings of all kinds, where everyone's a goofy-looking animal and you talk to others with spatial audio in a customizable browser-based 3D world. It's now open to everyone with a demo that doesn't require registering, and it's free to create a world of your own. [more inside]
It's an interactive explanation of a channel vocoder. (Homer Dudley's original vocoder from the 1930s was channel vocoder.) It let you perform all of the steps that the vocoder goes through to analyze a music signal and a voice signal and synthesize them together.
After two and a half years of work -- recording and editing some 300 hours of audio with dozens of actors -- I have started releasing the second season of my audio drama. Nineteen episodes (including a forthcoming seven part epic set between 1994 and 2023), 120 speaking roles, and I wrote nearly a thousand pages. The series involves parallel universes, wisecracking demons, revolutionaries, exuberant aliens fond of American nostalgia, and, above all, an examination of love and empathy.
A monthly soundscape mix of music and other sounds for every year - starting in 1853 and working my way slowly towards the present day. Also a monthly radio show on Cambridge 105fm where I talk about the recordings to give some context. Mixes have so far reached 1918 (plus 2016 and 2017) and the radio show is up to 1907.
It has been five years since Michael Brown, Jr. was killed in Ferguson, Mo. I work at a local news organization, so my colleagues and I cover this story and its effects daily — not just when it happened or on anniversaries. But anniversaries can be an appropriate time for reflection. So for this anniversary, we found people who have been living #Ferguson these past five years, and let them tell their own stories. We built this site to share them.
I've listened to podcasts ever since it required hunting Podcast Pickle, and love meta shows which repurpose audio (the defunct Ask Mr Biggs comes to mind) so I spent a few hours trying out an idea I've had rattling around: I've spliced together all 2018 welcomes/thanks of guests on CBC:s Quirks and Quarks podcast into a 15 minute thing. [more inside]
Listen to two guys talk about the Wire every two weeks. We often get off on tangents, most interesting, some not.
It's not a long "read" — I work at a radio station after all — but more of a long "listen and look". On the anniversary of the killing of Michael Brown, we remember what it felt like on those first few confusing days in August, melting into grief, anger and distrust coming from all directions, followed by the uneasy anticipation of the grand jury's decision, the chaotic nights after no true bill and then, finally, some progress. A project by me and my colleagues at St. Louis Public Radio.
Play an 80s synth in your browser with 106.js, a MIDI-enabled emulation of the Roland Juno-106 synthesizer. Chrome/Desktop only. Github repo here.
Wavelist lets you create and discover playlists of interesting podcast episodes. The first few playlist topics include marriage equality, the earthquake in Nepal, Sina Weibo and social networking in China, and weird birds.
Hey there! I make Rack Extensions for Reason, a music-making program from Propellerhead Software. (Rack Extensions are plug-ins for Reason, sort of like VSTs that you'd use for other DAWs.) I've made 10 so far. My latest is Chip64, an emulation of audio chips from vintage computers and video game consoles. There are some audio examples too.
Put Metafilter Music in your pocket! With MetaFilter Radio for iOS, you can listen to songs posted to Metafilter Music, hear episodes of the Metafilter Podcast, and retrieve and play your playlists, all from your iPhone or iPod touch. Requires iOS 7 or higher. [more inside]
GIFbites is a regular podcast and tumblr on/about/for animated GIFs. Each episode lasts no more than 15 seconds. [more inside]
The Conversation is an 8-month long conversation about the future, based on the hypothesis that at certain times in history the established systems (of economics, society, religion, etc.) cease answering the important questions. At these moments, we posit, a conversation takes place which tears down the old and builds anew the systems of thought for the next era. Our questions: Are we in one of those moments? If so, is the conversation taking place? Is the conversation even possible in this age of siloed knowledge and specialization. We're trying to find it or, perhaps, to start it. [more inside]
I wanted one slim bit of what last.fm provided - logging every time I play a song, and making that data easy to visualize. This is that bit - an open source OSX menubar app that outputs JSON.
"At The Tone" is the first comprehensive audio survey of NIST Radio Stations WWV and WWVH: two legendary shortwave radio broadcasters whose primary purpose is the dissemination of scientifically precise time and frequency. [more inside]
A site for my new band. The design is old skool, but the meat of the project is the 7 hours-worth of recordings on the "Listen" page. [more inside]
Fans of Lost, Twin Peaks and old time radio will enjoy the Post-Meridian Radio Players' newly completed series, "The Mask of Inanna". It's been called "a fascinating hybrid of a modern story with mystic overtones (reminiscent of Twin Peaks or the more fantastical tales of Stephen King), coupled with a beautifully-done 'show within a show' pastiche of classic old time radio". It's twelve episodes with months of production work per episode and a cast of over thirty. Please check us out! [more inside]
Audio Grains is dedicated to interesting Music Technology. Providing in-depth information and commentary topics ranging from equipment, software, audio theory, techniques and research. While leaning heavily on the technology side, we will also bring you music which uses technology in more interesting and creative ways. [more inside]
A tiny little browser-based app that will create a torrent of rain sound in response to the activity on a twitter search stream. This came out of Boston Music Hack Day 2011. This only works in Chrome, or bleeding-edge builds of Safari, due to its use of the brandy new web-audio API, which I didn't know existed until yesterday. [more inside]
My friend and I have created a 5-part audio drama about how a really bad Hollywood movie gets made. The story is told sequentially, in self-contained 30-minute pieces, from the perspectives of the writers, the producer, the script supervisor, the van driver, the assistant editor, and an audience member after the movie comes out. The first 30-minute piece took six months to produce (in our spare time with even sparer money), but I’m happy with how it turned out. We’re trying to decide if we want to record the rest of the episodes. I’m for it (as is the cast, etc.) but my friend’s not sure if people will be interested in listening to something like this. We would love any feedback you have to offer. Thank you! [more inside]
JP and Elah recently completed a year of weekly podcasts (46 episodes, we took a few breaks). Have a look at the episode archive and subscribe in iTunes, if you like. We're back with season 2 April 15th!
In a brand new venture, I'm hearkening back to the golden age of radio horror, when voices like Arch Oboler, Boris Karloff and Vincent Price ruled the airwaves. I've recorded two albums so far: MicroHorror, on which I read my own original short-short horror stories, and Legends of Horror, on which I read three classic stories by Edgar Allan Poe, Ambrose Bierce and H.P. Lovecraft in all-new dramatic performances. Streaming is free from Bandcamp, and downloads are pay-what-you-like. Dim the lights, curl up in your favorite coffin and open your mind to horror.