You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 786d1a2
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: fern/gpt-live/configuration.mdx
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -337,13 +337,13 @@ Send this body to `POST https://api.vapi.ai/call` with your Vapi private API key
337
337
338
338
Connect to `transport.websocketCallUrl` in the response. Send and receive binary audio frames containing signed 16-bit little-endian samples at 24 kHz. Convert your microphone audio to that format before sending it.
339
339
340
-
Set `24000` explicitly. The default WebSocket rate is too low for GPT-Live. Browser WebRTC and phone connections negotiate their own audio formats.
340
+
Set `sampleRate` to `24000` for GPT-Live. Browser WebRTC and phone connections negotiate their own audio formats.
341
341
342
342
Close the WebSocket to end the call, or have the assistant invoke `endCall`. Incoming JSON call-control commands are not supported by this GPT-Live connection. See [WebSocket transport](/calls/websocket-transport) for the general transport, applying the restrictions above.
343
343
344
344
## Review and monitor calls
345
345
346
-
Completed GPT-Live calls use Vapi's shared post-call processing. Configure the features you need, then inspect the results after a test call finishes.
346
+
GPT-Live supports post-call analysis, structured outputs, scorecards, Boards, and Monitoring. Configure the features you need, then inspect the results after a test call finishes.
Copy file name to clipboardExpand all lines: fern/gpt-live/limitations.mdx
+3-5Lines changed: 3 additions & 5 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -52,9 +52,7 @@ Early testing suggests some use cases no longer need the same divisions between
52
52
53
53
## Can I inject commentary or control speech during a call?
54
54
55
-
Customer-initiated commentary injection, live message injection, and live call-control commands are currently unsupported. Incoming text commands on the raw WebSocket connection are ignored.
56
-
57
-
Vapi uses OpenAI's commentary mechanism internally to pass reasoner results to the speaker. That doesn't expose `session.commentary.append` as a customer API. Use the supported [tool response flow](/gpt-live/configuration#connect-your-tools) to return results for requested work.
55
+
Commentary injection, live message injection, and live call-control commands are currently unsupported. The raw WebSocket connection supports audio streaming only. Use the supported [tool response flow](/gpt-live/configuration#connect-your-tools) to return results for requested work.
58
56
59
57
A caller can ask the assistant to slow down or explain differently, and you can prompt it to respond to those requests. That conversational behavior is separate from an application sending a control command.
60
58
@@ -80,13 +78,13 @@ If the caller corrects a date during a lookup, use the result that matches the u
80
78
81
79
## Can I run Simulations and Evals?
82
80
83
-
Voice Simulations are currently incompatible with GPT-Live. The simulation bridge uses 16 kHz audio and listen sockets. GPT-Live's raw WebSocket connection requires 24 kHz and doesn't provide listen sockets.
81
+
Voice Simulations are currently unsupported for GPT-Live.
84
82
85
83
Run voice scenarios through a [supported call connection](/gpt-live/configuration#connect-a-call). [Evals](/observability/evals-quickstart) can supplement testing of supported model decisions, but they don't establish GPT-Live's audio quality, timing, or interruption behavior.
86
84
87
85
## Recordings and post-call data
88
86
89
-
GPT-Live supports shared post-call analysis, structured outputs, scorecards, Boards, and Monitoring. Availability depends on your configuration and retained call data:
87
+
GPT-Live supports post-call analysis, structured outputs, scorecards, Boards, and Monitoring. Availability depends on your configuration and retained call data:
Copy file name to clipboardExpand all lines: fern/gpt-live/overview.mdx
+17-7Lines changed: 17 additions & 7 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -10,9 +10,25 @@ GPT-Live is an OpenAI voice model that can listen and speak at the same time. A
10
10
Vapi connects that conversation to your tools and phone numbers, and provides the call records, analysis, and monitoring around it.
11
11
12
12
<Note>
13
-
GPT-Live must be enabled for your Vapi organization. If you don't have access yet, [join the waitlist](https://vapi.ai/gpt-live-waitlist). We're admitting users from the waitlist every day.
13
+
**Private beta.**GPT-Live must be enabled for your Vapi organization. If you don't have access yet, [join the waitlist](https://vapi.ai/gpt-live-waitlist). We're admitting users from the waitlist every day.
14
14
</Note>
15
15
16
+
## In this guide
17
+
18
+
Start here for an overview, then choose what you need next.
19
+
20
+
<CardGroupcols={3}>
21
+
<Cardtitle="Design your assistant"icon="fa-light fa-comments"href="/gpt-live/design">
22
+
Plan speaker and reasoner roles, conversation flow, and speaking style.
Create an assistant, choose a voice, connect tools, and make a call.
26
+
</Card>
27
+
<Cardtitle="Limitations and FAQs"icon="fa-light fa-circle-question"href="/gpt-live/limitations">
28
+
Check supported features and what to consider before migrating.
29
+
</Card>
30
+
</CardGroup>
31
+
16
32
## How it works
17
33
18
34
A GPT-Live assistant uses a **speaker** for the spoken interaction and a **reasoner** for tasks that need tools or detailed procedures. **Delegation** is the speaker asking the reasoner to handle one of those tasks.
@@ -50,10 +66,4 @@ You also have more to design than the words alone. Set a voice, give the speaker
50
66
51
67
GPT-Live has a different set of supported features from Vapi's other voice architectures. Check [Limitations and FAQs](/gpt-live/limitations) if your application depends on squads, live call control, simulations, or a specific transfer flow.
52
68
53
-
## Get started
54
-
55
-
-**Explore the experience:** read [Design your assistant](/gpt-live/design) for architecture, conversation flow, and voice design.
56
-
-**Build a working assistant:** follow [Configure GPT-Live](/gpt-live/configuration#create-an-assistant) in the dashboard or API.
57
-
-**Assess a migration:** review [Limitations and FAQs](/gpt-live/limitations) against your current requirements.
58
-
59
69
For the underlying model, see OpenAI's [GPT-Live documentation](https://developers.openai.com/api/docs/guides/live). This section describes Vapi's integration and its supported configuration.
0 commit comments