Scalable, Efficient Processing and Analysis of Large Audio Datasets

This conference talk from the Audio Developer Conference (ADC) 2024 explores innovative approaches to processing and analyzing massive audio datasets. Learn how to handle terabytes of audio data efficiently using distributed computing with the Ray framework. Pawel Cyrta, an Applied Research Scientist with over 20 years of experience in audio technology and machine learning, demonstrates practical methods for scaling audio processing workflows, including feature extraction techniques like Mel-frequency cepstral coefficients and spectrogram analysis. Discover how to build robust, scalable data processing pipelines that can significantly accelerate audio analysis tasks for machine learning applications. The presentation provides actionable strategies for managing large-scale audio datasets, aggregating results, and deriving meaningful insights that can be applied to speech recognition, synthesis, and generative audio AI projects.