A token-based accent normalization pipeline using HuBERT discrete tokens, TTS-synthesized targets, and flow matching improves accent conversion over a frame-to-frame baseline.
Sub- jective evaluations on multiple English accents show signifi- cant improvements in speech naturalness, speaker similarity, and accentedness reduction
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Accent Normalization Using Self-Supervised Discrete Tokens with Non-Parallel Data
A token-based accent normalization pipeline using HuBERT discrete tokens, TTS-synthesized targets, and flow matching improves accent conversion over a frame-to-frame baseline.