TokenSE applies a Mamba architecture to discrete token prediction in a neural audio codec to enhance speech for cochlear implant users, showing better objective and subjective performance than baselines in noisy conditions.
The root- mean-square (RMS) level of all utterances was normalized to approximately 65 dB, and all audio files were sampled at 16 kHz
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
TokenSE: a Mamba-based discrete token speech enhancement framework for cochlear implants
TokenSE applies a Mamba architecture to discrete token prediction in a neural audio codec to enhance speech for cochlear implant users, showing better objective and subjective performance than baselines in noisy conditions.