Bridging the Gap: Integrating Pre-trained Speech Enhancement and Recognition Models for Robust Speech Recognition.
Kuan-Chen WangYou-Jin LiWei-Lun ChenYu-Wen ChenYi-Ching WangPing-Cheng YehChao ZhangYu TsaoPublished in: CoRR (2024)
Keyphrases
- speech recognition
- noisy environments
- speech enhancement
- noisy speech
- speech signal
- pattern recognition
- speech recognition systems
- speaker identification
- noise reduction
- hidden markov models
- speaker independent
- language model
- automatic speech recognition
- background noise
- pre trained
- spectral subtraction
- speech synthesis
- signal to noise ratio
- probabilistic model
- feature extraction