I'm confused, why don't we see the actual model re...
# support
s
heeeyyy I'm confused, why don't we see the actual model response in the Live test of an Assistant Call like, here it seems like we see the transcript of the audio response of the model that's odd? I mean sure its just for testing, but it seems to be something that needs a fix isn't it https://cdn.discordapp.com/attachments/1388648765723906200/1388648766122360944/CleanShot_2025-06-29_at_00.33.012x.png?ex=6861bf71&is=68606df1&hm=9c71e29cb94996ef6f577c47e7b6a3fd1889cfecc49b857e3f84f7ac5ce7fa32&
for example here it was supposed to give me an email address found in a json file, and it gave me some gibberish 'Synaptic is synaptic at synaptic a I dot I m dot g service account dot com.'
k
Vapi shows a transcript of the audio from TTS, not the raw model response, to see the actual model output, use Call Logs or set up a webhook to capture the assistant’s full text..
s
sure, I know, I get that it would simply be cleaner for us to be able to see to see the actual response from the model, but yeah!
v
If you would like to see that feature on our platform, feel free to add it to the feature request list here -->