zzw922cn/awesome-speech-recognition-speech-synthesis-papers

Curated repository of research papers spanning the speech processing pipeline from foundational signal processing methods (HMMs, temporal classification) through modern deep learning architectures (RNNs, CNNs, diffusion models). Covers emerging applications including text-to-audio generation, music modeling, and confidence estimation alongside traditional tasks. Organized by topic with direct links to paper PDFs, enabling researchers to trace the evolution of techniques from classical approaches to contemporary transformer and diffusion-based systems.

3,119 stars. No commits in the last 6 months.