SpecTNT: a Time-Frequency Transformer for Music Audio.
Wei Tsung LuJu-Chiang WangMinz WonKeunwoo ChoiXuchen SongPublished in: CoRR (2021)
Keyphrases
- audio signals
- music score
- music information retrieval
- music scores
- audio features
- audio signal
- audio recordings
- automatic music genre classification
- music genre classification
- music retrieval
- audio content
- speech music discrimination
- signal processing
- music collections
- digital audio
- content based music retrieval
- audio files
- polyphonic music
- fuzzy logic
- digital music
- musical instruments
- genre classification
- frequency domain
- wavelet transform
- power transformers
- hidden markov models
- visual information
- multimedia
- music recommendation
- acoustic features
- visual features
- fault diagnosis
- speaker identification
- cross modal
- partial discharge
- wavelet packet
- audio visual
- genetic algorithm
- feature set
- power system
- digital video
- fourier transform