Long tts message

Development
1

Many TTS providers have a limit on the message length that is allowed to be converted.
Right now it simply uses

self.send(self.player, PlayWavData(self.wav_data))

do they simply get played one after each other by the player?
-> I could use

for wav in self.wav_data:
    self.send(self.player, PlayWavData(wav))

with self.wav_data being an array of the splitted text converted to speech, or would I need to combine them and send them all in one call (I would prefer to send them without combining them, but I was not sure whether it would play them one after each other or might play multiple in parallel)

2

wav_data is the entire WAV file that the TTS system has generated. None of the Rhasspy TTS providers support splitting messages up right now. That needs to be handled at the sytem that’s making TTS calls into Rhasspy.