blog > gpt-live-1-1-upgrade

How we upgraded call4me to GPT-Live-1.1

The model swap was one line. Checking that phone calls didn't get worse was the rest of the evening.

OpenAI released GPT-Live-1.1 to the API on November 19, 2026, and call4me placed every phone call on it from 8:25 PM Pacific the same day. Changing the model took one line of configuration. The rest of the work was checking what a phone call depends on: the six voices customers choose from, how long a session lasts, how the caller answers an automated phone menu, and a deploy that doesn't cut off a call in progress.

call4me moving its phone calls from gpt-live-1 to gpt-live-1.1

What changed in GPT-Live-1.1

gpt-live-1.1 uses the same /v1/live/sessions endpoint as gpt-live-1. OpenAI says it transcribes speech more accurately and keeps a more consistent accent and voice across languages. Active session time still costs $0.05 a minute. gpt-live-1.1 model page

The changelog says to use gpt-live-1.1 in an existing Live session configuration and lists no breaking changes. The same entry adds standby, which pauses inference while nobody is speaking and bills that time at a lower rate. OpenAI API changelog

Two details mattered for a phone service:

The change itself

call4me's voice worker reads the model from one variable. This is the whole change to the voice configuration:

   "vars": {
-    "VOICE_MODEL": "gpt-live-1",
+    "VOICE_MODEL": "gpt-live-1.1",
     "BACK_OFFICE_MODEL": "gpt-5.5"
   },

Everything else about the session stays the same: 8 kHz G.711 μ law audio from Telnyx, the voice the customer picked, and a gpt-5.5 Responses backend that presses keypad digits, asks the user questions and ends the call. The voice architecture article shows that configuration in full, and the commit is public on GitHub.

Every voice still starts

Customers choose from six voices: marin, cedar, gleam, meridian, vesper and willow. The voice table in OpenAI's GPT-Live docs doesn't list every one of them, so we checked each voice on the new model instead of assuming.

We opened one gpt-live-1.1 session per voice with our phone audio format and played in a synthesized restaurant greeting. In all six, the session.started event echoed back gpt-live-1.1, the requested voice, audio/pcmu and 8000 Hz, and the caller answered the greeting with its line.

Sessions still end at two hours

GPT-Live closes a session two hours after it starts. A call that waits on hold can outlast that, so call4me replaces an expiring session mid-call with a new one that receives the call so far. Every 1.1 session we opened reported an expires_at exactly 7,200 seconds after it started, the same as gpt-live-1, so that handover needed no change.

Transcripts get the transcription upgrade

The transcript you get back after a call is built from GPT-Live's own transcript events for both sides of the line. We don't run a separate speech recognizer. If 1.1 transcribes more accurately, call transcripts and the summaries written from them improve with it.

A phone menu test before switching

The test we ran is an existing evaluation built from FedEx's automated phone menu, which listens for speech instead of keypad tones. Speech menus are strict: an answer padded with "uh" or "one moment" is often heard as no answer at all, and the menu asks again. The evaluation synthesizes FedEx's prompts word for word, streams them into a GPT-Live session at phone pace as 8 kHz μ law audio, and scores each answer on what such a recognizer can accept. The session gets the same instructions and backend a real call gets, and no phone call is placed.

We ran it three times on gpt-live-1.1 and compared it with the four runs on gpt-live-1 from November 18, which used the same caller instructions:

What the menu needs gpt-live-1 (Nov 18) gpt-live-1.1 (Nov 19)
The reason for calling in one sentence, no filler not recorded 2/3
A bare yes or no to a confirmation 4/4 3/3
A bare yes, or the key, when it offers both not recorded 2/3
The tracking number on the keypad, no words still failing 0/3
A bare no after it misheard 2/4 3/3
The tracking number keyed in when it says "say or enter" not recorded 3/3
A bare no to "are you the shipper" 3/4 3/3
A request for a person at a dead end, no filler still failing 1/3

These are small samples from two different days. On the rows where we have both numbers, 1.1 matched or beat gpt-live-1, so there was no regression on what we measured. It doesn't show 1.1 is better at phone menus.

The two weak rows were already open problems on gpt-live-1. On the keypad row, 1.1 twice read the digits aloud ("one, two, three...") instead of pressing them, and once said it was entering the number without pressing anything. A menu that accepts spoken digits would take those. The evaluation counts them as misses because the caller's instructions say numbers go in on the keypad, where a speech recognizer can't mishear them.

Deploying without dropping a call

Phone calls run in a separate Cloudflare Worker from the website, so a site deploy never touches a call. The voice worker only deploys when no call is active: the deploy script waits for zero active calls, then turns new calls away until the new version is serving. A call turned away in that window isn't charged and is told when to retry.

After the deploy we checked the version Cloudflare was serving. Its VOICE_MODEL was gpt-live-1.1.

What we haven't measured yet

Try it on your own calls

Nothing changes for call4me users: the same voices, profile and tools now run on gpt-live-1.1. If you want your coding agent to make phone calls, add call4me to your agent and give it a task with its boundaries:

Call the pharmacy and ask whether my refill is ready. If it isn't, ask when it will be and whether they need anything from me. Report what they said.

ai voice / gpt live / phone calls

comments

read next

An AI assistant calls a business for you and brings back the result
How to set up an AI that makes phone calls for you
Your AI assistant can call a business, talk to whoever answers and bring the result back to you. Here's how to connect it, what to ask for, and how to tell when the job is actually done.
OCT 7 / Nick Khami loading counts
OpenAI realtime voices on a real phone call, four voices compared
OpenAI realtime voices on a real phone call: marin vs cedar vs gleam vs meridian
Hear marin, cedar, gleam and meridian in five languages, then inspect repeated calls captured at the receiving end. We test dates, times, reference codes and corrections, with the recordings and transcription disagreements included.
SEP 30 / Nick Khami loading counts
An n8n AI Agent node uses the MCP Client Tool to reach call4me, places a phone call, and reports the outcome
How to make phone calls from n8n with the MCP Client Tool
n8n's MCP Client Tool turns any remote MCP server into tools for an AI Agent node. We pointed it at call4me, ran the workflow, and let it call Xfinity customer service. Here is the workflow, the settings that matter for long calls, and what the call actually got.
OCT 13 / Nick Khami loading counts

Nick Khami @skeptrunefounder of call4me, the phone for your AI agent. a y combinator founder who has raised $3.5M in venture capital and was head of engineering at a series b a16z startup. previously founded Trieve. email me@call4.me.

get new posts by email / loading reader count
no tracking, no spam; one email per post, unsubscribe in one click. or use the RSS feed. posts marked paid need a supporter subscription (monthly, cancel any time).

comments

loading comments

older: Mediacom customer service number: how we reached a person
all posts / archive / atom feed