Session I Monday, May 14 8:30 - 13:00 hr Room B
Watermarking & Internet Audio
Chair: Fons Bruekers, Philips Research, Eindhoven, The Netherlands
8:30 hr I-1
Design and
Implementation of a Real-Time Audio Service using MPEG-2 AAC and Streaming
Technology
Jose Soler Lucas, Tae-jin Lee & Jin-woo Hong
ETRI, Taejon, Republic of Korea
During the last years a number of efforts have been done
towards audio streaming. But among them, little has been done on MPEG-2
Advanced Audio Coding. Here we present the process by which we have built an
AAC streaming service over RTP to be used in Internet. The result obtained is
quite satisfactory and fulfils all our prospects.
Paper 5343
9:00 hr I-2
Audio Gateway
Concept: Migration from Conventional Dedicated to IP Audio Transmissions
Detlef Wiese
Consultant, Hallbergmoos, Germany
How to cover the migration from point-to-point
transmissions for contribution, distribution and other audio communication via
E1, T1 or ISDN to IP-audio with low cost access to Virtual Private Networks.
Quality of Service by ISPs can open the door to high quality professional audio
links for production, archiving, broadcasting, reporting and many other
applications. The proposed audio gateway concept is discussed regarding its
flexibility in bit rate, quality, delay time, bandwidth as well as
possibilities of new applications, such as connection events to Internet, ISDN
to Intranet, format conversion and others.
Printed Paper not available
9:30 hr I-3
The Internet
Applications of Embedded Metadata within Audio Watermarked Red Book Compact
Discs
Sigmund Rothschild
University of Colorado, Denver, CO, USA
Audio watermarking permits metadata to be encoded
directly into digitized audio and video program materials. The University of
Colorado at Denver�s CAM Records will embed metadata by watermarking their
current CD project using AudioTrack�s technologies. ABX double blind testing
will be used to evaluate the watermarking�s perceptual transparency. With the
encoded CD, CAM Records will implement several applications of the embedded
metadata within the context of their CD release, emphasizing the development of
Internet strategies and uses of the embedded metadata.
Printed Paper not available
10:00 hr I-4
Combined
Compression/Watermarking for Audio Signals
Frank Siebenhaar, Christian Neubauer & J�rgen Herre
Fraunhofer IIS, Erlangen, Germany
Perceptual audio coding has become a customary technology
for storage and transmission of audio signals. Audio watermarking enables the
robust and imperceptible transmission of data within audio signals, thus
allowing valuable information to be attached to the content, such as song
title, name of the composer and artist or property rights related data. This
paper describes a new concept for simultaneous low bit rate encoding and
watermark embedding for audio signals. In particular, the advantages of this
combined technique over separate steps of encoding and watermark embedding are
discussed (i.e. encoding of watermarked PCM audio signals or watermarking of
existing bit streams). Experimental results obtained from a first
implementation of an extended MPEG-2/4 AAC encoder are shown.
Paper 5344
10:30 hr I-5
Robust,
Multi-Functional, and High Quality Audio Watermarking Technology
Michiel Veen (1), Werner Oomen (2), Fons Bruekers (1), Jaap
Haitsma (1), Ton Kalker (1) & Aweke Negash Lemma (2)
Philips
Research, Eindhoven, The Netherlands
Philips ASA
Labs, Eindhoven, The Netherlands
Based on existing technology used in image and video
watermarking, we have developed a robust, multi-functional and high-quality
audio watermarking technique. The embedding algorithm operates in the frequency
domain, where the magnitudes of the Fourier coefficients are slightly modified.
Watermark detection relies on cross-correlation techniques, in which not only
the presence of a watermark, but also its payload is detected. Experiments
demonstrated that for this particular watermark, objective (ITU-R BS.1387) and
subjective (ITU-R BS.1116) audio quality measures correlate fairly well.
Combined analyses of the perceived audio quality and robustness indicated that
specific watermark parameters may be optimized for different applications.
These range from copy management (limited information capacity, high
robustness, and very high audio quality) to broadcast monitoring (intermediate
to large information capacity, intermediate robustness, intermediate to high
audio quality).
Paper 5345
11:00 hr I-6
A Compatible Family
of Bitstream Watermarking Schemes for MPEG-Audio
Christian Neubauer,
Ralph Kulessa & J�rgen Herre
Fraunhofer IIS, Erlangen, Germany
Audio watermarking enables the robust and
imperceptible transmission of data within audio signals. Among the many
possible applications of this technique, a number of scenarios require direct
embedding of watermarks into a compressed signal representation. Such bitstream
watermarking systems enable e.g. on-the-fly embedding of
transaction specific data at the time of content delivery via the Internet.
This paper extends previous work on bitstream watermarking for MPEG-2/4
Advanced Audio Coding (AAC) towards a compatible family of watermarking schemes
for MPEG-1/2 Layer-3 (MP3), MPEG-2/4 AAC and uncompressed audio. Regardless of
the format used during data embedding, the same watermark extractor can be
employed to recover the embedded message. Both the underlying concepts and
relevant experimental results for these schemes are described.
Paper 5346
11:30 hr I-7
Full-Text Indexing of
Very Large Audio Data Bases
Frank Kurth & Michael Clausen
University of Bonn, Bonn, Germany
We present a system for indexing of and index-based search
in PCM-based audio material. Given a short excerpt of a waveform signal as a
query, the index returns all pieces in a database containing that waveform.
Additionally, the precise position of the waveform within those pieces is
returned. The indexing method is robust against several signal processing
operations such as lossy compression or addition of noise. Indexing of a test
database consisting of approx 10 GB of audio data results in an index of size
16 MB. Response times to queries of lengths of about one or a half of a second
are only fractions of a second.
Paper 5347
12:00 hr I-8
Development of a
Distance-Learning Environment, using Database Driven Dynamic Web Pages -
Application for Digital Audio Internet Courses
Christos Sevastiadis,
Costas Rizakos, George Kalliris, Charalampos
Dimoulas & George Papanikolaou
Aristotle University of Thessaloniki, Thessaloniki, Greece
The current work focuses on the development of a web-based
user-friendly environment, for the design and management of distance learning
courses. Its objective is to promote the simple web service to a dynamically
refashioning, multimedia enabled tool in the hands of the tutor, offering
advanced learning facilities to the trainees. Based on the core of previous
work and the incorporation of contemporary technologies, such as a database
server and a web portal application server, a system capable to manage numerous
users and courses, to accommodate many new services and to provide new means of
communication has been brought up. The system is currently being evaluated
through a digital audio distance-learning course that is offered via the
Internet to the students of our department.
Paper
5348
12:30 hr I-9
Online Archives - the Pressure of Metadata
Siegbert Herla
Institut f�r Rundfunktechnik (IRT), Munich, Germany
For an economic realization of future online
archives working in a global network structure the right course shall be set.
Therefore already during the digitization and capturing process of single sound
carrier archives the registration of relevant metadata attention shall be paid
besides the re-recording of sound and video data. The description of sound and
video content, their technical quality, cue sheet and status information are
also important as a unique material identifier for a world-wide unique
identification of source material. In addition the continuous gathering of
metadata shall accompany the audio and video material on the road. A platform
independent container such as the BWF file is used to gather all the helpful
information on the way.
Paper 5349
