Overlapped speech detection using long-term spectro-temporal similarity in stereo recording.
Bo XiaoPrasanta Kumar GhoshPanayiotis G. GeorgiouShrikanth S. NarayananPublished in: ICASSP (2011)
Keyphrases
- long term
- short term
- detection algorithm
- computer vision
- detection accuracy
- object detection
- speech recognition
- detection rate
- speech signal
- false alarms
- automatic detection
- stereo matching
- detection method
- false positives
- real time
- stereo images
- face detection
- early vision
- noisy environments
- three dimensional
- audio visual
- voice activity detection
- binocular disparity
- text to speech
- stereo pair
- hidden markov models
- multi camera
- image pairs
- stereo vision
- depth map