| tech,15-8-H90-1060,bq | target speaker </term> and combined by <term> | averaging | </term> . Using only 40 <term> utterances </term> | #17185 Each reference model is transformed to the space of the target speaker and combined by averaging . |
|
|
|
|
| tech,15-8-H90-1060,bq | target speaker </term> and combined by <term> | averaging | </term> . Using only 40 <term> utterances </term> | #17185 Each reference model is transformed to the space of the target speaker and combined by averaging . |
