Support for UI Agent Settings #56
Description
I'm curious if there will ever be support for UI based agent Settings. It looks like there is an internal endpoint that the UI uses to do things like set voice settings for the agent globally rather than only in a published version in an environment. The endpoint is
https://dialogflow.clients6.google.com/v2beta1/projects/[agentId]/locations/global/agent:updateSettings?alt=json&key=[apiKey}
Here's an example payload:
{"agentSettings":{"enableSmalltalk":false,"enableValidation":false,"enableSpeechRequestContext":true,"enableAgentWideKnowledgeConnector":false,"knowledgeBaseNames":[],"knowledgeConfidenceAdjustment":0,"textToSpeechSettings":{"enableTextToSpeech":true,"outputAudioEncoding":"OUTPUT_AUDIO_ENCODING_LINEAR_16","synthesizeSpeechConfigs":{"en":{"speakingRate":1,"pitch":0,"volumeGainDb":0,"effectsProfileId":[],"voice":{"name":"en-US-Wavenet-F","customVoice":{}}}}}},"updateMask":"enableStackdriverLogging,enableSmalltalk,enableSpellCorrection,disableAutomaticTraining,enableValidation,enableSpeechLogging,enableSpeechRequestContext,endpointerSensitivity,enableSentimentAnalysis,enableAgentWideKnowledgeConnector,enableCustomSpeechModels,knowledgeBaseNames,knowledgeConfidenceAdjustment,textToSpeechSettings"}
i don't see any protobuf files for agentSettings...
Other functions that we need in the client library are the telephony integrations, especially the one click Google one.
Right now these things are blocking us from being able to programmatically provision agents and have them fully set up without having to go into the UI and click buttons. This is driving us towards seeking alternate solutions that do enable us to do that. But we would love to stick with dialogflow...