[ 8. Februar 2026 ]

DEGEM News – FWD – [ak-discourse] Einladung zum Forschungskolloquium am 10.02.2026

Von: ak, sekretariat via ak discourse
Datum: Thu, 5 Feb 2026
Betreff: [ak-discourse] Einladung zum Forschungskolloquium am 10.02.2026

Liebe Freunde der Audiokommunikation,

am kommenden Dienstag, den 10.02.2026, um 18:00 Uhr im Raum EN 324 präsentiert Nikolaj Wolff seine Masterarbeit zum Thema

„De-amping: removing amplifier and distortion effects from guitar recordings“

Dazu möchten wir Sie herzlich einladen. Eine Kurzzusammenfassung finden Sie am Ende dieser E-Mail.

Der Zoom-Link für unsere auswärtigen Gäste lautet:

tu-berlin.zoom.us/j/65356423211<https://tu-berlin.zoom.us/j/65356423211?pwd=33q4fHr0fhhV2WCxYDNL2LM4S2hVe2.1>

Meeting-ID: 653 5642 3211
Kenncode: 141962

Mit herzlichen Grüßen
Kirill Andreev & Stefan Weinzierl
__________________

Nikolaj Wolff: De-amping: removing amplifier and distortion effects from guitar recordings

De-amping—the process of recovering the direct input signal from a distorted electric guitar recording—is a challenging task with applications in music production, sound restoration, and music information retrieval. While only a handful of recent studies have directly addressed de-amping, the current state of the art employs two-stage neural architectures that first denoise spectrograms and then reconstruct the waveform using a neural vocoder. This study investigates whether deep neural networks (DNNs) can reverse the linear and nonlinear transformations introduced by guitar amplifiers and effects. Several DNN architectures are evaluated using signal-based metrics and a perceptual listening test. The best results are achieved by a two-stage system combining the BigV-GAN v2 neural vocoder with the Cascaded Gaze spectrogram denoiser, which demonstrates strong robustness, especially in zero-shot scenarios on unseen data. The relationship between listening test results and signal-based metrics is examined to assess how accurately these measures represent subjective audio quality.
__________________

Kirill Andreev
Er/ihm – he/him
Anrede/Form of address: Herr – Mr.

Sekretariat
Office

Technische Universität Berlin
Fakultät I / Inst. f. Sprache u. Kommunikation
FG Audiokommunikation / Sekr. EN-8

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

Einsteinufer 17, 10587 Berlin
GERMANY
sekretariat@ak.tu-berlin.de
+49 (0)30 314-22236

https://www.tu.berlin/ak