SECP: A Speech Enhancement-Based Curation Pipeline For Scalable Acquisition Of Clean Speech.
Adam SabraCyprian M. WronkaMichelle MaoSamer L. HijaziPublished in: CoRR (2024)
Keyphrases
- speech enhancement
- noisy speech
- noisy environments
- speech signal
- noise reduction
- speech recognition
- signal to noise ratio
- hidden markov models
- single channel
- vocal tract
- linear prediction
- background noise
- automatic speech recognition
- wiener filter
- additive noise
- smoothing algorithm
- edge preserving
- multi channel
- lossless compression
- signal processing
- speech synthesis
- language model
- spectral subtraction
- information retrieval