IEEE

Mel-Spectrogram Feature Extraction for Raw Acoustic Classification of Electric Motor Performance

Abstract: This paper presents a non-intrusive method for electric motor performance classification using acoustic signals and Mel-spectrogram-based feature extraction. A dataset of 900 WAV files, ...
IEEE

Environmental Sound Classification Using Feature Fusion of MFCCs, Mel-spectrogram, and Chroma

Abstract: Environmental sound recognition enables automated systems to interpret and respond to diverse acoustic environments, enhancing applications in safety, monitoring, and smart technologies.
GitHub

LM2A:lyrics and motion to audio with diffusion model

Data sets: Songs 2020, 2021, 2022: Each data set contains all lyrics, segmented lyrics (sliced), motion parameters' JSON files, and audio files. Downloaded BigVGAN This code is used to preprocess a ...
GitHub

Deep learning scaffold for decoding audible Morse code into text on macOS with GPU acceleration (MPS). The project is set up for training a CTC acoustic model on log-mel ...

PyTorch wheels for Apple Silicon/Intel macOS include MPS support. Verify the device at runtime (the training script does this automatically): There isn’t a widely used public Morse-audio benchmark; ...