Audio AI Training in Warsaw, Poland

This  training introduces participants to the foundations and applications of Audio AI, focusing on audio processing, sound classification, speech recognition, and generative audio using machine learning. The training blends theory with hands-on practice using open-source tools and models.

Fundamentals of Audio AI and Recognition

Introduction to Audio AI

• What is Audio AI and how it's used
• Applications: speech recognition, sound detection, generative audio
• Tools and libraries: Librosa, PyDub, SpeechRecognition, torchaudio

Working with Audio Data

• Audio file types, formats, and sampling rates
• Loading, visualizing, and segmenting audio
• Extracting features: MFCC, spectrograms, chroma, zero crossing

Audio Classification Basics

• Building a simple sound classification model
• Training on environmental sounds or musical instruments

Speech Recognition with Pretrained Models

• Intro to speech-to-text
• Using libraries like Vosk, Whisper, or Wav2Vec2

Generative Audio and Advanced Applications

Noise Reduction and Audio Enhancement

• Identifying and removing background noise
• Signal processing filters and neural noise suppressors

Text-to-Speech and Voice Synthesis

• Understanding TTS pipelines
• Tools: Google TTS, Coqui TTS, ElevenLabs (overview)

Intro to Audio Generation with AI

• Music generation basics (MusicLM, Jukebox overview)
• Creative use cases: podcasting, sound design, storytelling
• Limitations and ethical considerations

Final Exercises and Wrap-Up

• Summary, best practices, and further learning paths