Synthetic Speech Detection Based on the Temporal Consistency of Speaker Features.
Yuxiang ZhangZhuo LiJingze LuWenchao WangPengyuan ZhangPublished in: IEEE Signal Process. Lett. (2024)
Keyphrases
- temporal consistency
- speech recognition
- false positives
- speaker verification
- audio visual
- speaker identification
- detection algorithm
- speaker recognition
- automatic speech recognition
- mel frequency cepstral coefficients
- noisy environments
- detection accuracy
- speech signal
- image features
- feature vectors
- constraint satisfaction problems
- post processing
- object detection
- optical flow
- speaker diarization
- face recognition