Hyperaud.io is a service that allows people to transcribe and word-align audio and video. The resulting Hypertranscripts can be used to remix and share media.
Audiovisual content containing the spoken word becomes much more accessible and discoverable when we couple it with a transcript. Making the transcript interactive adds another layer of accessibility, shareability, and remixability. Using speech-to-text and alignment services, we can create word-level-timed interactive transcripts, or “Hypertranscripts”. Hyperaud.io is intended as a complete ecosystem to allow the transcription, correction and alignment of transcripts, so they can be remixed into new forms using existing text-oriented paradigms. We promote media-literacy by removing the barrier to creating video-essays and other new forms of audiovisual expression. Because remixing is non-destructive, this allows people to remix remixes into new and expansive pieces.
Hyperaudio is a codebase as well as a service and has been used in projects such as Palestine Remix (palestineremix.com), Studs Terkel Radio Archive (studsterkel.wfmt.com) and heavily inspired commercial projects such as Trint (trint.com)
Acknowledgments: Mozilla
Knight Foundation
Team Information: The Hyperaudio Team, Contact:
Mark Boas
[email protected]
Members:
Laurian Gridinoc
Piotr Fedorczyk
Mark Panaghiston
Matteo Spinelli
Dan Schultz
Mark Boas