Vapi Comments

•Created by subsequent-cyan on 12/6/2025 in #support

Test Suite vs Evals

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

3 replies

VVAPI

•Created by rare-sapphire on 12/6/2025 in #support

New Public key creation for every new Assistant

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

5 replies

VVAPI

•Created by fair-rose on 12/6/2025 in #support

call transfer

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

5 replies

VVAPI

•Created by harsh-harlequin on 12/6/2025 in #support

Neha Voice Not Selectable in Assistant Configuration

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

3 replies

VVAPI

•Created by spiritual-aqua on 12/6/2025 in #support

schedulePlan Not Working

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

3 replies

VVAPI

•Created by deep-jade on 12/6/2025 in #support

Cannot publish any of the openai realtime models

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards,
Vapi
Customer Support Team

4 replies

VVAPI

•Created by stormy-gold on 12/4/2025 in #support

GPT Realtime model failing to respond back to user with tool call response

Thanks for reporting this issue. We're looking into it and will get back to you with an update once we do some testing

5 replies

VVAPI

•Created by continuing-cyan on 12/4/2025 in #support

Inefficient algorithm for MiniMax TTS

you can also try disabling chunking completely:

{
  "voice": {
    "provider": "minimax",
    "cachingEnabled": true,
    "chunkPlan": {
      "enabled": false
    }
  }
}

{
  "voice": {
    "provider": "minimax",
    "cachingEnabled": true,
    "chunkPlan": {
      "enabled": false
    }
  }
}

9 replies

VVAPI

•Created by harsh-harlequin on 12/1/2025 in #support

Question about difference between Transcript(Final) and transcript

The transcript server message sends live, partial transcription as the caller speaks. It updates continuously and may change as recognition improves. The Transcript (Final) server message only sends finalized, high-confidence text after each utterance is fully processed. It won’t change once delivered and is best for storing or post-processing. If you need the transcript of the entire call, use the end-of-call report, which includes the full call transcript along with every other call data point.

8 replies

VVAPI

•Created by stormy-gold on 12/4/2025 in #support

Metrics Page

Thank you for reporting this issue. We were able to reproduce it on our side and will be discussing it with the team. We’ll provide an update soon. Best,
Vapi Support

4 replies

VVAPI

•Created by manual-pink on 12/4/2025 in #support

Which server message to listen to know when the assistant ends to speak

Thanks for the detailed question — here’s the most reliable way to detect when the assistant has finished its LLM output. At the moment, Vapi does not emit a dedicated “final LLM token” event. The events that exist (model-output, conversation-update, speech-update) can fire multiple times within a single turn or occur only after TTS finishes, which is why you're seeing inconsistency. The recommended approach is: - Use conversation-update events if you want the completed turn with full message history. - Use speech-update if you want to detect when the assistant begins speaking its final response for that turn (the started event fires once per turn). - If you need to trigger logic exactly when the last LLM token is generated, the best option today is to implement your own buffer tracking on the server side and treat the last model-output event before the next speech-update as the true end of the LLM turn. This combination is currently the most reliable way to detect end-of-turn behavior before TTS begins.

8 replies

VVAPI

•Created by like-gold on 11/26/2025 in #support

Outbound Call to NZ Numbers not working

Glad you were able to get this resolved. We’ve closed this ticket, but your input is very valuable to us. Please take a moment to complete the survey, which should appear in a couple of days—your feedback directly helps us improve our support.

10 replies

VVAPI

•Created by deep-jade on 11/25/2025 in #support

What is the TTL for Voice Caching?

Thanks for the question — here’s how the voice caching works: TTL for cached voices:
There’s no fixed TTL. ElevenLabs audio is cached on our side without an expiration and is only removed when our Redis cache evicts it using LRU, or when it’s manually cleared internally. Enforcing expiration:
There isn’t a way to programmatically expire cached voices through the API. Cache management happens entirely on our backend. Best workaround:
Modify any synthesis parameter (stability, style, voice settings, etc.). This generates a new cache key and forces fresh synthesis, which is the most reliable way to bypass the existing cache without disabling caching.

4 replies

VVAPI

•Created by conscious-sapphire on 11/25/2025 in #support

integration

Hi, could you please provide a call id where you experienced this issue? We would like to review the logs to identify the cause.

3 replies

VVAPI

•Created by rare-sapphire on 11/25/2025 in #support

Squad calls the tool without filling the path parameter.

Hi, could you please provide a call id where you experienced this issue? We would like to review the logs to identify the cause.

3 replies

VVAPI

•Created by deep-jade on 12/3/2025 in #support

First Message not invoked when using custom tts server

Hey! To help track down this issue, could you share: - The call ID - When exactly this happened (the timestamp) - What response you expected to get - What response you actually got instead This would really help us figure out what went wrong!

12 replies

VVAPI

•Created by helpful-purple on 11/28/2025 in #support

Twilio Number Setup - An error occurred in this call: call.start.error-get-transport

Glad you were able to get this resolved. If anything else comes up or you need further assistance, we’re here to help.

72 replies

VVAPI

•Created by modern-teal on 12/3/2025 in #support

call start error

Thanks for your patience while we dug into this. Looking at call 019ae38f-47d1-7336-867d-e5fb2b1244ef, we’re seeing the following internal error on our side: 🔴

 09:33:30:127 Couldn't initiate outbound call transport and queue for call 019ae38f-47d1-7336-867d-e5fb2b1244ef. Error: Couldn't Create Twilio Call. Twilio Error: Authenticate

This means Vapi wasn’t able to authenticate with Twilio when trying to start the outbound leg. It’s not related to your assistant configuration, but to the telephony transport step. A few things to check on your side while we continue investigating: 1. Twilio credentials in Vapi - Re-save your Twilio Account SID and Auth Token in the provider/phone number settings (in case the token was rotated or invalidated). 2. Twilio account status - Confirm in the Twilio console that the project is active, has sufficient balance/credit, and outbound calling is allowed for your region and numbers. 3. From number configuration - Make sure the Twilio number you’re using for outbound is still valid, SMS/voice-enabled, and not restricted.

6 replies

VVAPI

•Created by ambitious-aqua on 10/26/2025 in #support

LangSmith observability provider

Glad you were able to get this resolved. If anything else comes up or you need further assistance, we’re here to help.

5 replies

VVAPI

•Created by foreign-sapphire on 11/21/2025 in #support

Support Request – Transfer Call Not Working (ISDN / Zadarma SIP Trunk)

Hi Gerald, Thanks for the detailed technical summary. Very helpful. Zadarma may have limitations or requirements around handling REFER-based call transfers. It would be best to contact Zadarma support directly and confirm: - Whether they fully support SIP REFER for blind transfers on your specific trunk - Whether any special configuration is required on their end - Whether number length or formatting affects REFER handling - Whether blind transfer flows (REFER → INVITE) are supported for ISDN destinations Because Vapi only sends a REFER during blind transfer, and we can see in your trace that the REFER was sent correctly and accepted, the next step definitely lies with Zadarma’s processing of that REFER. If you get any technical feedback from them and want us to take a second look, feel free to share it here — happy to help interpret it. Best regards,
Vapi Support

8 replies