I'm Sorry for Your Loss: Spectrally-Based Audio Distances Are Bad at Pitch.
Joseph TurianMax HenryPublished in: CoRR (2020)
Keyphrases
- multimedia
- acoustic features
- distance measure
- signal processing
- audio visual
- visual information
- distance function
- dissimilarity measure
- audio stream
- speaker identification
- cross modal
- multimedia information
- visual data
- neural network
- digital video
- emotion recognition
- euclidean distance
- image processing
- distance matrix
- audio signals
- audio files
- music score
- database