Spectral cross-correlation features for audio indexing of broadcast news and meetings.
Masahide YamaguchiMasaru YamashitaShoichi MatsunagaPublished in: INTERSPEECH (2005)
Keyphrases
- cross correlation
- broadcast news
- speaker diarization
- story segmentation
- template matching
- speaker identification
- frequency domain
- mutual information
- feature vectors
- audio visual
- automatic speech recognition
- feature space
- information retrieval
- high dimensional
- image features
- machine learning
- video search
- spoken document retrieval
- video sequences