Skip to main content

Table 9 Relative improvement depending on the content and style layers in cases of Google cloud-based evaluation using 10% UME-ERJ subset

From: Accent modification for speech recognition of non-native speakers using neural style transfer

Style layers Content layers RI(CER)
1–10 6–12 40%
1–8 8–12 36.1%
1–10 10–12 38.3%
1–5 5–12 29.4%
1–4 4–12 21.1%