Audio Processing for the Internet of Things (AIoT)2.5 ECTS
(englische Bezeichnung: Audio Processing for the Internet of Things)
(Prüfungsordnungsmodul: Audio Processing for the Internet of Things)

Lectures:

• • Audio Processing for the Internet of Things
    (Vorlesung, 2 SWS, Nils Peters, Wed, 16:15 - 17:45; from 21.4.2021 to 7.7.2021)

Empfohlene Voraussetzungen:

It is recommended to finish the following modules before starting this module:

Signale und Systeme I (WS 2020/2021)
Digitale Signalverarbeitung (WS 2020/2021)
Signale und Systeme II (SS 2020)

Inhalt:

The course focuses on audio and speech processing algorithms within the context of the Internet of Things (IoT).

• Foundation (history, components, current challenges)

• Overview of relevant wireless protocols (bandwidth, range, latency, spectrum)

• Audio device synchronization (NTP, PTP, device orchestration, acoustic wireless sensor networks, asynchronous and event-driven audio sampling)

• Acoustic Sensing for Voice User Interfaces (keyword spotting, speech recognition, speaker verification, anti-spoofing)

• Acoustic Scene Detection (event detection, scene classification, anomaly detection, sound tagging, blind reverb estimation)

• Sound Creation (text-to-speech, sound generative networks)

• Data-over-sound (sound-beacon, watermarking, acoustic fingerprint)

• Privacy in IoT (edge vs. cloud processing, secure signal processing, federated learning, differential privacy, audio encryption)
  

Lernziele und Kompetenzen:

The students will be able to

• understand the principles, key components, and current in IoT

• know the differences between various wireless transmission protocols and can give recommendations based on the IoT use case

• understand the differences of edge- and cloud-based audio signal processing

• understand algorithmic strategies to enhance privacy in IoT use cases

• understand the algorithmic components in a voice user interface

• understand state-of-the art methods for detection and classification of acoustic scenes and events

• learn and apply algorithms to transmit data via acoustic signals

• quantify the impact of latency in audio networks and apply strategies for acoustic device synchronization
  

Literatur:

Recommendations for each topic are given during the lectures

Organisatorisches:

In this course, we require a good knowledge of deep learning techniques, machine learning, and pattern recognition as well as a strong mathematical background. Furthermore, we require a solid background in general digital signal processing and some experience with audio, image, or video processing.

It is recommended to finish the following modules (or having equivalent knowledge) before starting this module:

• Lecture Deep Learning

• Digitale Signalverarbeitung

• Statistische Signalverarbeitung

• Sprach- und Audiosignalverarbeitung
  

Weitere Informationen:

Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan:

1. Advanced Signal Processing & Communications Engineering (Master of Science)
   (Po-Vers. 2020w | TechFak | Advanced Signal Processing & Communications Engineering (Master of Science) | Gesamtkonto | Technical Electives | Audio Processing for the Internet of Things)

Studien-/Prüfungsleistungen:

Audio Processing for the Internet of Things (Prüfungsnummer: 45221)

Prüfungsleistung, mündliche Prüfung, Dauer (in Minuten): 30, benotet, 2.5 ECTS

Anteil an der Berechnung der Modulnote: 100.0 %

Prüfungssprache: Englisch

Erstablegung: SS 2021, 1. Wdh.: WS 2021/2022

1. Prüfer:

Nils Peters