[ 11. Februar 2024 ]

DEGEM News – FWD – [ak-discourse] Einladung zum Forschungskolloquium am 13. Februar 2024

Von: Andreev, Kirill
Datum: Fri, 9 Feb 2024
Betreff: [ak-discourse] Einladung zum Forschungskolloquium am 13. Februar 2024

Liebe Freunde der Audiokommunikation,

am kommenden Dienstag, den 13. Februar 2024, 16:00 Uhr im Raum EN 324 präsentiert Ali Diab seine Masterarbeit zum Thema

„Beyond Functionality: A Typological Exploration of Distinct Architectural Strategies for Fostering Creative Atmospheres in Recording Studios“.

Im Anschluss präsentiert Marius Seipel seine Masterarbeit zum Thema
„Improving speech enhancement models with neural post-processing“.
Dazu möchten wir Sie herzlich einladen. Eine Kurzzusammenfassung finden Sie am Ende dieser E-Mail.

Der Zoom-Link für auswärtige Gäste lautet:


Meeting-ID: 656 0395 9784

Kenncode: 20231009

Mit herzlichen Grüßen
Kirill Andreev & Stefan Weinzierl

Ali Diab: Beyond Functionality: A Typological Exploration of Distinct Architectural Strategies for Fostering Creative Atmospheres in Recording Studios

The evolution of music and its performance spaces have historically been intertwined. However, with the invention of sound storage and playback devices, a transformative shift occurred, giving rise to a new era marked by experimentation at the intersection of music and technology. Recording studios emerged as a significant adaptation to this technological progress, prompting an exploration into whether this typology was solely a consequence of technological advancements or if spatial considerations played a role. Examining successful music studios reveals a persistent influence of spatial quality and experience. While acoustical standards provide diverse solutions, architectural ideas offer opportunities to craft unique spaces with captivating atmospheres. Thus, this research hypothesizes that in today’s recording studios, often confined to rigid templates, it’s crucial to recognize the importance of creating recording spaces that go beyond mere functionality. These spaces should possess the ability to evoke an atmosphere that enhances the musicians‘ experiences, and help them reach their creative zenith.

Marius Seipel: Improving speech enhancement models with neural post-processing

Real-world audio is often degraded due to noise, distortion or low-quality equipment and therefore not suitable for high-quality speech applications. In this work, a state of the art speech enhancement and restoration network is utilized and further improved with the help of post-processing methods. In addition to a 1-D convolutional waveform-to-waveform network, a DSP normalization was used as a traditional comparison. The listening test results show small improvements in speech quality and overall quality for the post-processing and indicate that the standalone waveform-to-waveform post-processing network works.

Kirill Andreev




Technische Universität Berlin

Fakultät I / Inst. f. Sprache u. Kommunikation
FG Audiokommunikation / Sekr. EN 8 Raum 321

Faculty I / Institute for Language & Communication
Audio Communication Group / Office No. EN 8 Room 321

Einsteinufer 17, 10587 Berlin
+49 30 314-22236