Closed grigio closed 4 years ago
The easiest way is to POST WAV data to /api/speech-to-text
or /api/speech-to-intent
. Rhasspy will automatically convert it for you and return the transcription or recognized intent.
If you want to stream audio continuously, there are a few options: via MQTT and using GStreamer are probably the easiest.
Is there an official API to stream or send the voice audio to Rhasspy? Is https://github.com/synesthesiam/rhasspy/pull/143 the raccomanded way to do it?