Decoupling Recognition and Transcription in Mandarin ASR

Yuan, Jiahong; Cai, Xingyu; Gao, Dongji; Zheng, Renjie; Huang, Liang; Church, Kenneth

doi:10.1109/asru51503.2021.9688053

Cited by 7 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Grapheme/phoneme conversions that convert text/character sequence into phoneme sequence, and regenerate text/character sequence from the phoneme sequence that is generated from source speech [87].…”

Section: Connections To Other Methodsmentioning

confidence: 99%

“…• Automatic Speech Recognition. Some works [78,87,89] in automatic speech recognition first transform target character sequence into phoneme sequence using grapheme-to-phoneme (G2P) conversion, and use an acoustic model to predict phoneme sequence from source speech sequence, and then use another model to generate character sequence from predicted phoneme sequence. • Text Generation.…”

Section: Applications Of Regeneration Learningmentioning

confidence: 99%

See 1 more Smart Citation

Regeneration Learning: A Learning Paradigm for Data Generation

Xu¹,

Qin²,

Bian³

et al. 2023

Preprint

View full text Add to dashboard Cite

Machine learning methods for conditional data generation usually build a mapping from source conditional data X to target data Y . The target Y (e.g., text, speech, music, image, video) is usually high-dimensional and complex, and contains information that does not exist in source data, which hinders effective and efficient learning on the source-target mapping. In this paper, we present a learning paradigm called regeneration learning for data generation, which first generates Y (an abstraction/representation of Y ) from X and then generates Y from Y . During training, Y is obtained from Y through either handcrafted rules or selfsupervised learning and is used to learn X → Y and Y → Y . Regeneration learning extends the concept of representation learning to data generation tasks, and can be regarded as a counterpart of traditional representation learning, since 1) regeneration learning handles the abstraction (Y ) of the target data Y for data generation while traditional representation learning handles the abstraction (X ) of source data X for data understanding; 2) both the processes of Y → Y in regeneration learning and X → X in representation learning can be learned in a self-supervised way (e.g., pre-training); 3) both the mappings from X to Y in regeneration learning and from X to Y in representation learning are simpler than the direct mapping from X to Y . We show that regeneration learning can be a widely-used paradigm for data generation (e.g., text generation, speech recognition, speech synthesis, music composition, image generation, and video generation) and can provide valuable insights into developing data generation methods.

show abstract

Section: Connections To Other Methodsmentioning

confidence: 99%

Section: Applications Of Regeneration Learningmentioning

confidence: 99%