Hi all, I work a lot on language education in virtual worlds and have been trying to intergrate 'text to speech' in-world with chatbots. The problem I encounter is that the playback of a mp3 (converted externally from text) can be done by parcel media or parcel audio, but badly. Every viewer responds differently on this. The best result I have untill now is on the Firestorm viewer with parcel MEDIA, that plays back the sentence but with a delay of 5 seconds which is really unnatural. When I use parcel AUDIO (streaming music) it starts right away BUT that viewer has an audio fade-in of a few seconds so I can only hear the last part of the sentence spoken. It all depends on the viewer.
So my question is... Would it be possible to create some new way of scripting (some additional osPlayExternalAudio) to play external mp3 files from within an object. With some http request? That would eliminate the viewer problems. b.t.w. the spoken mp3 has a different name every time it is loaded. Are there other users that also think this would be a great new option or am I the only one? Let me know what you think or how you have solved this problem!!! Regards, Nick Zwart = = = = = = = = = = = = = = 3DLES Fluitekruidlaan 25 3925 SB Scherpenzeel Gld The Netherlands phone: +31 33 4611902 fax. +31 33 4612723 cell +31 6 10577016 e-mail: [email protected] skype: nick.j.zwart site: www.3dles.com = = = = = = = = = = = = = = _______________________________________________ Opensim-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/opensim-users
