Conference Schedule
During the conference, live A/V streams are available for the main track: High Quality and Low Bandwidth.
Remote participants are invited to join #lac2015 on irc.freenode.net, to be able to take part in the discussions, ask questions, and get technical assistance in case of stream problems.
Conference Material can be found on the Download Page.
Timetable Format: Plain List | Table | iCal | Printable Version
All times are CEST = UTC+2
Poster Session
11:30 Auphonic: web-based automatic audio mixing and post production system - Poster Presentation(60 min) Georg Holzmann » Location: Foyer
Auphonic is a Linux-based web service to automate audio post production and mixing using a combination of music information retrieval, machine learning and signal processing algorithms. Users upload their multi- or singletrack recordings and Auphonic will analyze the audio, applies all processing automatically and keeps learning and adapting to new data every day!
The system includes an Adaptive Leveler, which balances levels between speakers, music and applies dynamic range compression, Loudness Normalization to new broadcast standards (EBU R128, ATSC A/85, ...), a True Peak Limiter, Automatic Ducking, Noise Gate and Cross Talk Removal algorithms for multitrack recordings, Automatic Noise and Hum Reduction and much more.
Furthermore it can create multiple audio file formats with correct metadata and distribute all results to various destinations (FTP, SFTP, HTTP, WebDAV, YouTube, SoundCloud, Dropbox, Google Drive, S3, Archive.org, ...). The algorithms can be used directly in the Web Interface, with an API or in a separate desktop version.
The whole system is based on open source libraries and runs on a Linux system, most of our algorithms are implemented in Python/numpy/C.
We have introduced a freemium model to finance auphonic.com, which allows us to work on new algorithms every day.
The system includes an Adaptive Leveler, which balances levels between speakers, music and applies dynamic range compression, Loudness Normalization to new broadcast standards (EBU R128, ATSC A/85, ...), a True Peak Limiter, Automatic Ducking, Noise Gate and Cross Talk Removal algorithms for multitrack recordings, Automatic Noise and Hum Reduction and much more.
Furthermore it can create multiple audio file formats with correct metadata and distribute all results to various destinations (FTP, SFTP, HTTP, WebDAV, YouTube, SoundCloud, Dropbox, Google Drive, S3, Archive.org, ...). The algorithms can be used directly in the Web Interface, with an API or in a separate desktop version.
The whole system is based on open source libraries and runs on a Linux system, most of our algorithms are implemented in Python/numpy/C.
We have introduced a freemium model to finance auphonic.com, which allows us to work on new algorithms every day.
The schedule is a major guideline. There is no guarantee events will take place at the announced timeslot.
