Research Article
Utterance Clustering Using Stereo Audio Channels
Table 1
Mean z-scores of error rates on different methods.
| Data sets | Mono | mstack | hstack | Sumdif |
| With overlap | 0.2520 | 0.1861 | −0.1477 | −0.2905 | Without overlap | 0.2764 | −0.1419 | −0.0872 | −0.0473 |
|
|