Pith. sign in

As mentioned earlier, for narrowband speech, the spectral features represent information only from 0-4 kHz and the remaining 4-8 kHz are missing

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

eess.AS 1

years

2019 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

background 1

representative citing papers

Bandwidth Embeddings for Mixed-bandwidth Speech Recognition

eess.AS · 2019-09-05 · conditional · novelty 4.0

A bandwidth embedding, a small learned vector marking speech as narrowband or wideband, improves narrowband word error rate by 13% relative in a single mixed-bandwidth acoustic model without degrading wideband speech.

citing papers explorer

Showing 1 of 1 citing paper.

  • Bandwidth Embeddings for Mixed-bandwidth Speech Recognition eess.AS · 2019-09-05 · conditional · none · ref 3

    A bandwidth embedding, a small learned vector marking speech as narrowband or wideband, improves narrowband word error rate by 13% relative in a single mixed-bandwidth acoustic model without degrading wideband speech.