| W00-0602 | inherent uncertainty in spoken | language transcriptions | , which almost always contain |
| W12-1613 | a reliable resource for spoken | language transcription | ( Marge et al. , 2010 ) . Transcriptions |
| W06-2304 | canonical material such as spoken | language transcription | or e-mails . This technique , |
| D10-1072 | structured text or even spoken | language transcriptions | that are not organized into discrete |
| W00-0602 | orthographic differences , since spoken | language transcriptions | lack many features present in |
| D12-1070 | end-to-end WFST source to target | language transcription | and translation system with syntactic |
| C96-2192 | As noted earlier , the spoken | language transcriptions | contain many deviations fl ` |
| C96-2192 | lematic . First of all , spoken | language transcriptions | are typically produced ill a |
| W15-4914 | texts into written text and spoken | language transcriptions | , and , contrary to the other |
| H90-1020 | will be derived from the natural | language transcription | while listening to the spoken |
| D12-1070 | resource-poor language and resource-rich | language transcriptions | . This creates an integrated |
| W98-0208 | mechanisms for low quality spoken | language transcriptions | of broadcast stories . Other |
| W15-4727 | is compatible with the natural | language transcription | phase but impacts the structured |
| P09-1062 | to its independence from spoken | language transcription | , an error-prone and resource-intensive |
| W07-0604 | community standards for child | language transcription | present a disincentive for child |
