Is it possible to supply custom voice to Dialogflow integration?

[Kelvin_Lau]

Hi there,

We are implementing a Dialogflow voicebot. We wondered if it is possible to supply custom voice from a third party Speech synthesis engine. I saw resemble.ai is able to change the bot voice using SSML webhook and fulfilment. so just wondering if we could do the same for Genesys Dialogflow integration.

We are aware that Genesys does not support third party TTS. But since this is more on modifying the Dialogflow side, so we still want to try our luck

Regards,
Kelvin

[Anton_Vroon]

This is certainly something we would be interested in too.

However unless your third party does both STT and TTS then I doubt it will work. You set a single language option which needs to handle both parts.

I have an idea open to allow you to mix and match STT and TTS that might allow what you want, you can add your use case to this idea:
https://genesyscloud.ideas.aha.io/ideas/STA-I-13

[Kelvin_Lau]

I saw on resource center that it is possible to switch bot voice. Therefore I suspect it is viable by modifying the Dialogflow audio output , just not sure how the backend works with Genesys telephony integration

[Kelvin_Lau]

A little finding here...

I have tried the SSML response . It worked perfectly with Genesys telephony integration.

<speak> Hello! <break time = \"1s\"/>
<audio src=\"https://actions.google.com/sounds/v1/alarms/bugle_tune.ogg\">
</audio> 
<break time = \"1s\"/>
<prosody rate=\"slow\"> What dfo you think?</prosody>
</speak>

Looks like we just have to set up a webhook call to our TTS api and return the custom audio url in fufilment response, then that's it!