The interactive and non-linear aspects of video games bring forth the problem of repetition.In game sound, dialogue is particularly susceptible to repetition because of our hearing’ssensitivity to human speech. The most common way to avoid repetition in game dialogue isto record multiple takes of every phrase as to have several naturally varied samples, whereasthis study explores the possibility to create these natural variations by manipulation of pitchand/or time of one recorded phrase using a readily available software. A listening test wasconducted where 23 subjects rated the variation, naturalness and artifacts of a recordedspoken phrase where three altered versions manipulated by altering the pitch, timing and bothpitch and timing where compared to the original recording, examining which manipulationtype yielded the most variation while also considering the naturalness and artifacts. Arepeated measures factorial ANOVA and pairwise comparisons showed a significantdifference between all manipulation types regarding the three dependent variables. Theresults show that the pitch manipulated sample had the best compromise between perceivedvariation, naturalness and artifacts, and would therefore be the recommended method forcreating variation of recorded dialogue.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:ltu-74142 |
| Date | January 2019 |
| Creators | Timan, Jesper |
| Publisher | Luleå tekniska universitet, Institutionen för konst, kommunikation och lärande |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0022 seconds