0

Running on Ubuntu 16.04 on a X86 64 architecture. I am following the Google Assistant SDK for python guide : https://developers.google.com/assistant/sdk/guides/library/python/embed/run-sample Everything seems to work well except that I get no sound when I run the test. Hereafter the output :

ON_CONVERSATION_TURN_STARTED

ON_END_OF_UTTERANCE

ON_RECOGNIZING_SPEECH_FINISHED: {'text': 'TF1 21h'}

ON_RESPONDING_STARTED: {'is_error_response': False}

ON_RESPONDING_FINISHED

"TF1 à 21h" is the phrase I gave but I don't get any spoken answer. Does not seem to be a sound system issue since when I go to dialogflow console training/history there is no trace of the call, so I assume the call din't reach dialogflow. One more thing my dialgflow app is in French. Any idea how I could find where is the issue ?

Thierry Donnadieu
  • 29
  • 5

1 Answers1

0

The Google Assistant SDK gives you the ability to programmatically make audio requests to the Google Assistant. In ordinary usage you won't be using Dialogflow at all.

All requests to the Assistant can be seen by looking at your activity so you can see if there is an audio response.

The audio requirements for the Google Assistant SDK library include ALSA, which may or may not be available on your computer.

Nick Felker
  • 11,536
  • 1
  • 21
  • 35
  • Many thanks, very useful, didn't realize that this library gives you access to the overall Google Assistant not to your specific app. I could also use directly Dialogflow api but I need a speech to text api and I couldn't find one at Google besides using google assistant SDK. Any suggestion on your side ? – Thierry Donnadieu Apr 04 '18 at 08:50
  • Dialogflow has a similar gRPC API for [sending text or audio](https://dialogflow.com/docs/reference/api-v2/rpc/google.cloud.dialogflow.v2beta1#detectintentresponse) although I'm personally not familiar with it. – Nick Felker Apr 04 '18 at 18:23