Reported by jteh on 2015-02-03 00:56
A lot of potentially nice functionality is not possible (or is at least ridiculously painful) with our current speech framework. Speech needs a pretty big refactor to allow for such things. It needs to support:
- Callbacks which are called at a requested point in the speech output
- Callbacks which are called when a given synth finishes speaking and when overall speech has stopped
- Speech commands in control/format field speech so that control and formatting info can be indicated by things other than just text
- Priority output so that important messages can interrupt what is being spoken and/or be spoken after the next utterance without losing lower priority utterances already sent
Unfortunately, some of this is going to break backwards compatibility, but I think it's worth it in this case.
Blocking #905, #3188, #3286, #3736, #4089, #4233, #4433, #4874, #4966, #5026, #5104
Reported by jteh on 2015-02-03 00:56
A lot of potentially nice functionality is not possible (or is at least ridiculously painful) with our current speech framework. Speech needs a pretty big refactor to allow for such things. It needs to support:
Unfortunately, some of this is going to break backwards compatibility, but I think it's worth it in this case.
Blocking #905, #3188, #3286, #3736, #4089, #4233, #4433, #4874, #4966, #5026, #5104