search

met4citizen / TalkingHead

Talking Head (3D): A JavaScript class for real-time lip-sync using Ready Player Me full-body 3D avatars.
MIT License
349 stars 107 forks source link

can i use msedge tts for lip sync? #64

Closed 24wemy closed 1 month ago

met4citizen commented 1 month ago

I'm not sure. In addition to audio, the TalkingHead class requires word-level timestamps for accurate lip-sync and the MSEdge TTS appears to be a very limited version of Microsoft Azure TTS. However, it does include a wordBoundaryEnabled option in its request headers. If setting that flag to true works, and the server actually delivers word-level timestamps, then you might be able to make it work.

24wemy commented 1 month ago

I'm looking for a free TTS, because I don't have access to Google TTS or Microsoft Azure, can you help me?

met4citizen commented 1 month ago

The TTS you suggested doesn't seem to have any official API documentation, which implies that it is intended for Microsoft's internal use only. Moreover, the projects that use it do so by pretending to be Edge browsers, which most likely violates Microsoft's API policies. -- If you can find some legal document stating that their API can be used by anyone for free, I'm happy to help. Otherwise, no.