Music Asset Management Solutions
Professional tools for organizing and managing large music collections
For labels, publishers, libraries, music houses and production organizations; One Llama Media has the tools to streamline your workflo, effectively mine your archives and manage any type of digital music assets.
One Llama's Music Asset Management System Features
Web Services
Whether you requirements are for in-house archive management, or distributing you music collection to a global online audience, One Llama has built all of its software tools to ensure rapid prototyping and easy website integration.
Audio Analysis
Overcome the "cold-start" problem with One Llama's proprietary audio analysis tools. The system can extract over 150 features from samples as brief as 5 seconds, in up to 4 times real-time. Whether or not you have metadata to start with, or if you want to supplement existing metadata, this best-in-class tool can help.
Machine assisted Classification
Using both acoustic and semantic analysis techniques, One Llama's tools can accelerate the process of accurately classifying music collections of almost any size. Using unique machine learning techniques, the system can be "trained" to recognize a customized taxonomy, identify similar tracks, and automatically sort them into categories. An easy to use web interface makes it simple for non-technical personnel to validate and "train" the system. New categories or tags can be added dynamically and the entire collection updated in near real-time.
Audio File Preparation
Preparing your files for digital distribution can be deceptively complex and time consuming. Metadata must be embedded for various formats in different ways. The One Llama system can ingest audio assets, and import data in almost any CSV format and batch embed into AIFF, WAV, BWAV and MP3 file formats to almost any industry standard including Soundminer and Netmix.
One Llama Music Asset Management Services
Taxonomy Engineering
Some of the world's leading experts in music information systems and semantics can assist you with perfecting your classification schema so that it is future-proof and usable. This is the first and most important step in building any Music Asset Management System.
Metadata Creation
No machine based system can complete the task of music metadata creation. Our network of highly skilled musicologists can be your personal outsourced team for adding comprehensive metadata to almost any set of tracks in the right style and format to suit almost any need. We can develop a workflow pipeline to meet a wide range of deadlines and track volumes. Our editors are selected from academic programs in information sciences and musicology and are usually graduate students.
Software Engineering
We can customize our code base to suit a wide variety of needs. Please send us your spec or RFP for a needs analysis and quote.
Audio File Preparation Utility
Embed detailed metadata in a variety of audio file types based on data drawn from any CSV or Excel spreadsheet. Import tools are provided to assist the user in configuring the tool used to import existing dumps of collected metadata, facilitating easy integration of the tool into their file preparation pipeline. The embedded metadata format is optimized for display in the majority of Digital Audio Workstation products, media libraries, and media players.
The simple user interface enables relatively non-technical staff to conduct batch operations on music repositories of almost any size.
Technical specifications
Operating systems supported:
- Mac OS X
- Windows XP and Vista
- Linux
Audio file types and metadata containers supported:
- MP3
- ID3v1.1
- ID3v2.3
- Full metadata, lyrics and artwork
- ID3v2.4 (on request)
- Soundminer
- Netmix (version 4+)
- Broadcast WAVE (BWAV or BWF)
- BEXT
- Soundminer, including artwork
- Netmix, including artwork
- AIFF
- ID3v2.3
- Soundminer
- Netmix (version 4+ - reading not currently supported, but expected soon)
The audio files produced are optimized for display in the following applications:
- MP3 and AIFF files
- iTunes
- Windows Media player
- All other MP3 players/libraries utilizing the ID3v2.3 standard
- Soundminer
- Netmix
- Broadcast WAVE (BWAV or BWF)
- Soundminer
- Netmix
- Digital Audio Workstation (DAW) products such as: Pro Tools, Nuendo, Avid, Final Cut Pro and all other DAW products supporting the BEXT/BWAV standard
Additional Features:
- Import arbitrary CSV or Excel spreadsheets into the tool and map your metadata to fields that will be embedded in the audio files.
- Supports multiple configurations (e.g. different formats or import settings for different sub-libraries, or for distribution to different clients or locales).
- Import settings may be saved and reused later.
- Maintain existing folder trees or restructure your collections by creating customized folder trees based on the content of metadata fields (for example, you could organize files by Library -> Sublibrary -> Album/release -> Artist etc.).
- Embed lyrics and artwork.
- Both batch and UI interactive modes supported.
- Optional metadata reports (for e.g. which fields are most often unpopulated, what proportion are populated, etc.)
Future Planned Release Roadmap
- Import wizard and settings manager
- Command Line (batch) interface
- Metadata reports
- Develop installer with dependency checking (Java 1.5 or greater with MacRoman and US-ASCII text encoding support - default Java install on Windows is currently missing MacRoman)
Anomaly Detection
Audio Mining Technology
One Llama has developed a state of the art suite of audio and generalised signal data mining and search tools. These tools have been successfully applied to the complex field of music search and classification and offer significant facilities for the classification, search and detection of anomalies within any audio or generalised signal sequence. Further, One Llama's powerful approaches to the learning of signal or anomaly types can facilitate the rapid development and deployment of solutions in new or as yet untested fields.
One Llama's current capabilities include:
Sequence Event detection and Annotation
Given an arbitrary signal or audio stream the technology can identify distinct events and be taught label the individual events or whole streams (based on the events detected).
Real-time Monitoring
Perform real-time analysis and issue informational or alarm signals on the detection of certain events or changes in their characteristics (e.g. breathing rate change).
Active-learning
State-of-the-art One Llama active-learning systems can be used to rapidly develop and refine systems to classify or label new data types and classes in an easy to use interface - allowing subject matter experts to impart their knowledge to the system with ease.
Bespoke Signal Processing Solutions
One Llama's music and generalised audio mining technology has been successfully applied to the detection and classification of events from a number of domains including music, vocal sound, environment sound, bird song and breathing sounds. The modular nature of One Llama's DSP, Machine-learning and interface software allows for the development of bespoke, optimized solutions in new domains requiring tailor-made analyses.
The One Llama research team also has significant experience in the analysis of signals from other domains such as neurological scans, text documents and process monitoring data. Hence, One Llama is confident in being able to provide tailor-made solutions to the analysis of any suitably informative signals.
Methods of Deployment
- Web-based interfaces to system monitoring, training and reporting tools. Web-services enable simple integration with existing IT systems.
- Provision of DIY system customisation tools or consultancy based analysis & optimisation services.
- Solutions provided on multiple scales including PC, remote compute service and embedded hardware
Potential Applications
The applications of One Llama's technology that have been envisaged include:
- Breathing sound monitoring
- Industrial Process monitoring
- Surveillance video navigation
- Smart Baby monitoring
- Consignment or payload monitoring
- Audio-based intrusion detection
- Animal call and bird song transcription Human vocal sound detection
- Automated audio channel optimization
