Skip to main content
Back to voice channels

Set up Realtime LLM Voice channel

Realtime LLM Voice is a voice channel that handles live voice conversations through a single provider-native speech session. It supports Realtime LLM models (such as OpenAI Realtime and Gemini Live) through the Kore.ai Voice Gateway (KoreVG) or a Bring Your Own SIP connection. Unlike the STT → LLM → TTS pipeline that Pipeline Voice uses, a provider-native realtime model owns the live audio conversation. KoreVG handles telephony, connection routing, bounded session state, traces, and the call lifecycle.
Studio also provides in-product guidance. On the Realtime LLM Voice channel page, click Setup instructions to open a step-by-step panel that covers connection creation, provider and phone routing, call quality validation, and security, latency, and reliability boundaries.

Create a Realtime LLM Voice channel

To create a Realtime LLM Voice channel, go to Deployments > Channels > Voice > Realtime LLM Voice > New connection and complete the following steps in order.
1

Set up the connection

On the New Realtime LLM Voice Connection page, provide the following fields.
2

Create the connection

Click Create.
Create the connection before you configure voice details. The connection record must exist before Studio can provision a phone number and save provider-specific realtime settings.

Configure the Realtime LLM Voice channel

Once the channel is created, open it from the Realtime LLM Voice connections list. The connection has three tabs: Settings, Speech, and Testing. The Settings tab holds the voice provider and call routing. The Speech tab holds realtime orchestration, the S2S provider, and advanced voice behavior. The Testing tab holds the shareable softphone link.
1

Choose a call routing option (Kore.ai Voice Gateway)

On the Settings tab, under Realtime LLM Voice Settings, set Voice Provider to Kore.ai Voice Gateway or Bring Your Own SIP. If you choose Kore.ai Voice Gateway, set the following under Call routing. Choose how calls reach this agent. Enable Soft Phone for browser testing, and add a phone number (DID) when the channel needs a real phone number.Keep KoreVG as the managed telephony gateway unless your platform owner has approved another provider. Confirm that the realtime provider you plan to use is available for your tenant and environment.
If you select Phone Number (DID), complete the next step. If you selected Voice Provider: Bring Your Own SIP instead, skip to step 3.
2

Configure the phone number (DID only)

  • Set Provider to Twilio.
  • Toggle Inbound calls (routes calls from this number into the voice agent).
  • Toggle Outbound calls (allows this number for campaign dial‑out).
  • Select a Country, Number Type, and optional Area Code.
  • Choose Search Available Numbers to list matching numbers.
Buy or assign a Twilio‑backed number from the saved connection’s settings.
  • Confirm the number is bound to this connection before activation.
  • After assignment, the number appears in the Phone Number field.
  • Select Remove to unbind the number from the connection.
3

Configure SIP settings (Bring Your Own SIP only)

If you chose Bring Your Own SIP, the Call routing options are replaced by a SIP Configuration section on the Settings tab. Set the following fields.
4

Set SIP header trust and save the connection settings

At the bottom of the Settings tab, for either voice provider, set the following.Click Save Changes to apply the settings.
5

Configure Realtime LLM Voice settings

On the Speech tab, set the following fields.
  • Realtime orchestration: Select how runtime coordinates the conversation flow and business tool execution for realtime voice interactions. Realtime orchestration determines whether the realtime model or runtime controls conversation progression and business actions, such as bookings, orders, and payments.
    • Changes to the orchestration mode apply only to new realtime sessions. Calls already in progress continue using the mode that was active when the session started.
    • Existing channel configurations without a configured orchestration mode continue to use Legacy behavior.
  • Response wording owner: Select who generates the wording of responses for realtime voice conversations. This setting is available only when the Realtime orchestration mode is set to Runtime Owned - canonical turn.
    • Runtime text (recommended): Runtime generates the response text. This option provides consistent response generation.
    • Provider-grounded voice: Uses the realtime voice provider to generate spoken responses while runtime continues to control conversation flow and business actions. This option is available only for Gemini Live. For unsupported or unsafe conversation turns, it automatically falls back to runtime-generated text responses.
  • Semantic convergence: Semantic Convergence determines how the platform coordinates with the realtime voice provider during a voice interaction. Controls whether realtime voice turns use the coordinated utterance runtime. This setting is available only when the realtime orchestration config is set to Legacy mode. The Effective at runtime line below the setting shows the mode currently in effect.
  • S2S Provider: The realtime speech-to-speech model the channel uses. Select an approved realtime provider and model, and match its audio format, locale, voice, and tool-calling capabilities to the agent. Select one of the following: Amazon Nova Sonic (S2S), Azure OpenAI Realtime, Grok Realtime (S2S), Google Gemini Live, or OpenAI Realtime. Each option shows which credentials it uses:
  • Realtime model connection: Shown for every S2S provider except Amazon Nova Sonic, which authenticates via Voice Service credentials instead. Selects the project-pinned realtime model this provider uses. It shows Use project realtime model until a workspace admin adds one. The Source line below the dropdown shows where the model comes from, and the deployment uses this exact model connection. To add one:
    1. Configure the model under Voice Services, then add it to the model catalog as an LLM provider.
    2. In the project, go to Settings > Models and add the model you configured at the workspace level. See Models.
  • Provider-specific settings: After you select an S2S provider, a configuration section for that provider appears. Expand the provider below to see its settings.
    Under Amazon Nova Sonic Configuration, set the following fields.Pending tool fillersTurn on Pending tool fillers to let Nova speak a brief hold phrase when a project tool is still running. Fast tools finish without a filler.Coordinator turn fillerTurn on Coordinator turn filler to let Nova speak a short hold phrase during convergence-enabled coordinator turns while Runtime decides and executes the turn.Advanced voice behaviourThe Guarded flow delivery settings are channel defaults for guarded realtime flows only. They apply when Realtime orchestration is set to Runtime guarded, and your values are kept when it’s set to another mode. An agent or step can override each value in its ABL. Changes apply from the next call.Also configure the Advanced voice behaviour settings that apply to all providers, described after the provider list.
    Under Azure OpenAI Realtime Configuration, set the following fields.Turn detectionAlso configure the Advanced voice behaviour settings that apply to all providers, described after the provider list.
    Under Grok Realtime Configuration, set the following fields.Also configure the Advanced voice behaviour settings that apply to all providers, described after the provider list.
    Under Gemini Live Configuration, set the following fields.Tool response schedulingRealtime turn fillerSeparate from project Voice Fillers. Applies only to Google or Vertex realtime S2S calls while semantic convergence is set to Enforce.Filler delivery safeguardsShared by pending-tool and realtime turn fillers.Voice activity detectionControls when Gemini starts listening to a caller turn and how long it waits before closing that turn.Caller transcriptionSilent-turn recoveryAdvanced voice behaviourAlso configure the Advanced voice behaviour settings that apply to all providers, described after the provider list.
    Under OpenAI Realtime Configuration, set the following fields.Turn detectionTune how OpenAI detects when speech starts and when the caller has finished speaking.Noise reductionThe provider filters the caller’s audio before it detects speech and before the model hears it, so background sound starts fewer false turns.
    Noise reduction removes noise, not other voices. Background talkers and echo of the agent still reach the model. This setting is separate from the gateway’s Noise isolation setting under Advanced voice behaviour.
    Caller transcriptionHow the caller’s speech is written down. This text is stored as the caller’s turn and is what the agent checks answers against.
    Never put a caller’s own details in Words to expect. The transcriber would then hear them from anyone.
    Realtime turn fillerApplies only to OpenAI Realtime calls while semantic convergence is set to Enforce. Off by default.Advanced voice behaviourAlso configure the Advanced voice behaviour settings that apply to all providers, described after the provider list.
  • Parallel caller transcription: For supported Kore.ai Voice Gateway (KoreVG) realtime channels, you can use a separate speech-to-text (STT) provider to transcribe caller audio while the realtime speech-to-speech (S2S) provider handles the conversation. The S2S provider continues to supply the agent’s response transcript.
  • Availability: Off by default. A platform administrator must enable the Realtime LLM parallel STT feature for your tenant. When enabled, use the Caller transcription card on the Speech tab to select a configured STT provider instance and the caller’s language.
  • Supported S2S providers: OpenAI Realtime, Azure OpenAI Realtime, Grok Realtime (S2S), and Google Gemini Live. Amazon Nova Sonic (S2S) isn’t supported.
  • Transcripts: Final transcripts from the STT provider are used for caller messages in the session transcript and conversation history, with one message per caller turn. Interim caller transcripts aren’t sent. Agent transcripts still come from the S2S provider. If the STT provider doesn’t return a final transcript within 3 seconds, the system falls back to the S2S caller transcript.
  • Troubleshooting: The voice_stt trace can show both the STT transcript (transcript) and the S2S transcript (s2sTranscript) for the same caller turn. They’re two values for one turn, not separate messages.
  • Billing: Caller audio goes to both the STT provider and the S2S provider, so review usage and billing for both.
The existing Caller wording setting controls which wording the agent runs a caller turn on. Itsis separate from selecting an STT provider for caller transcription. If you don’t see the Caller transcription card, ask your platform administrator whether the feature is enabled for your tenant. If the feature is revoked after configuration, the saved settings become read-only and new calls use S2S caller transcription.
  • Advanced voice behaviour (all providers): Optional behaviours for this channel. Each one is off, empty, or set to the earlier platform behaviour until you change it.
  • Voice transfer and survey capabilities: A read-only panel that shows how many of the transfer and survey capabilities Runtime evaluates as available for the concrete voice execution boundary (the execution surface is KoreVG realtime S2S), for example 11 of 11 capabilities available. It lists the following capabilities, each marked Available or not.
Click Save Changes on the Speech tab to apply the settings.
6

Save and confirm provider readiness

Resolve any missing provider credentials and model configuration before you test. After you save, confirm that Studio shows the phone number on the saved connection. Each tab has its own Save Changes button, so save on each tab you edit.
Don’t repoint shared KoreVG or softphone hosts. Phone registration, device call, Jambonz API, and SBC hosts are shared environment settings. Only a connection-specific callback may target a local tunnel during isolated testing.

Test and verify the channel

Open the Testing tab to verify the channel before going live.
  • Shareable softphone link: Generate a time‑limited link that lets anyone call the channel from a phone browser through the real gateway and view a read‑only transcript.
    • Links expire after 7 days and can be revoked at any time.
    • Active links shows links currently in use, or No active links for this channel.
    • Viewing active links requires read access; generating or revoking links requires edit permission.
  • Soft Phone dialer: From the Settings tab, select Soft Phone > Launch dialer to call the agent directly from your browser.
    • No DID purchase or telephony charges are required.
Verify and secure the Realtime LLM Voice channel Before you go live, validate the quality of realtime calls. Realtime quality depends on the slowest leg of the call, so test each of the following.
1

Preview with HTTPS voice

Generate a time‑limited preview, allow microphone access, start and stop a call, mute/unmute, and confirm chat remains usable if microphone or provider setup fails.
2

Run an isolated PSTN test call

Dial the provisioned number and validate greeting, bidirectional audio, tool execution, transfer/termination, caller hang‑up, and one stable Runtime session.
3

Measure latency budget

Record gateway connect time, provider session setup, input audio commit, first model audio, end‑to‑end turn p50/p95, jitter, and packet loss.
4

Test barge‑in and rapid turns

Interrupt a long agent response, speak immediately, and verify stale audio is cancelled, tool calls aren’t duplicated, and the next user turn is preserved.
5

Exercise failure scenarios

Deny microphone access, force provider disconnect, gateway reconnect, model timeout, tool failure, phone hang‑up, and capacity rejection. Confirm caller hears or receives the designed fallback.
Security, Latency, and Reliability Practices
  • Protect credentials: Treat realtime provider and carrier credentials as administrator‑managed secrets. Don’t copy API keys into connection identifiers, browsers, screenshots, or agent prompts.
  • Minimize overhead: Avoid unnecessary transcoding, buffering, proxy hops, and oversized tool responses. Set explicit connection and turn timeouts, and verify heartbeat behavior through every proxy.
  • Validate capacity: Check provider concurrency, KoreVG session limits, phone capacity, and backpressure before load tests. Capacity rejection must fail predictably, not degrade active calls.
  • Define fallback behavior: Specify caller‑visible handling for provider unavailability, timeout, safety rejection, tool failure, and transfer failure. Include rules for retry, reprompt, mode switch, or call termination.
  • Trace sessions: Capture call/session IDs, provider events, tool calls, turn latency, interruptions, errors, and termination reasons. Scrub audio, transcripts, and credential data from logs.

Manage an existing Realtime LLM Voice channel

Open a channel from Deployments > Channels > Voice > Realtime LLM Voice and select a connection. The connection header shows the display name, its status (for example, Active), and when it was last updated. The connection has three tabs. Changing settings on the Settings or Speech tab updates the live channel after you click Save Changes.

Troubleshoot Realtime LLM Voice

  • Calls to the number fail: an Active phone-routed connection without a provisioned number can’t receive production PSTN calls. Assign or buy a number, and confirm it’s bound to this connection.
  • The channel can’t start a session: resolve missing provider credentials and model configuration, then save the connection and test again.
  • Duplicate tool calls or stale audio after an interruption: repeat the barge-in test from Verify and secure the Realtime LLM Voice channel and check the traces for the affected call and session IDs. For the full validation checklist, click Setup instructions on the Realtime LLM Voice channel page.

Set up Pipeline Voice channel

Pipeline Voice is a voice channel that handles calls through separate speech-to-text (STT), LLM, and text-to-speech (TTS) stages, connected through Kore.ai Voice Gateway (KoreVG). Unlike Realtime LLM Voice, which uses a single speech-to-speech model, this pipeline lets you choose and configure the speech recognition (ASR) and speech synthesis (TTS) providers independently. KoreVG captures the caller’s speech through STT, sends the transcript to the deployed agent, and streams or plays the agent’s response through TTS. Each stage has its own provider, latency, failure, and observability boundary.
Choose your voice provider before you begin. It determines which call routing fields apply later in this setup.
Studio also provides in-product guidance. On the Pipeline Voice channel page, click Setup instructions to open a step-by-step panel that covers connection creation, speech pipeline and phone route configuration, validation of every voice phase, and security, queue, and reliability boundaries.

Create a Pipeline Voice channel

To create a Pipeline Voice channel, go to Deployments > Channels > Voice > Pipeline Voice > New connection and complete the following steps in order.
1

Set up the connection

On the New Pipeline Voice Connection page, provide the following fields. The setup instructions appear alongside the form while you fill it in.
2

Create the connection

Click Create. Create the connection before you configure the pipeline, because the saved connection owns the phone route and its STT, TTS, barge-in, no-input, and timeout configuration.
Configure the channel Once the channel is created, open it from the Pipeline Voice connections list. The connection has four tabs:
  • Settings holds the voice provider, call routing, and advanced voice behavior.
  • Speech holds speech recognition, speech synthesis, and conversation controls.
  • Deployment holds the environment binding.
  • Testing holds the shareable softphone link.
Resolve missing STT and TTS credentials before you preview or activate the channel. Speech providers are administrator-configured credential sets. See Voice Services.
Configure the voice provider and call routing
1

Choose a voice provider

On the Settings tab, under Pipeline Voice Settings, set Voice Provider to Kore.ai Voice Gateway or Bring Your Own SIP. Keep Kore.ai Voice Gateway selected unless your platform owner approves another route. KoreVG owns the Jambonz session, sticky gather, streaming TTS, interruption handling, and voice trace phases.
2

Configure call routing (Kore.ai Voice Gateway)

If you chose Kore.ai Voice Gateway, set the following under Call routing. Choose how calls reach this agent. Enable Soft Phone for browser testing, and add a phone number (DID) when the channel needs a real phone number.
If you enable Phone Number (DID), complete the next step. If you chose Bring Your Own SIP, skip to step 4.
3

Configure the phone number (DID only)

Set the following fields, then select Search Available Numbers to list matching numbers. Verify that the number you buy or assign is bound to this connection before you activate it.Until the connection has a provisioned number, the channel header shows a Needs phone provisioning badge, and the page shows the warning Configured without a phone number. Assign or buy a number before you route phone calls.
4

Configure SIP settings (Bring Your Own SIP only)

If you chose Bring Your Own SIP, the Call routing options are replaced by a SIP Configuration section on the Settings tab. Set the following fields.Select one of the following Transport options.Under Gateway options, select the options that apply to your carrier.
Outbound calls need a SIP termination URI. Add one, or turn off Outbound calls. Otherwise, the platform rejects Save Changes.
Toggle Enable E.164 (leading +) to prepend a leading + to outbound numbers sent to the carrier. It’s on by default, which international (full-E.164) numbers require. Turn it off only for local or national-format numbers, since the leading + otherwise misroutes them.
5

Configure advanced voice behavior

Under Advanced voice behaviour, set the optional behaviors for this connection. Select the section header to expand or collapse it. Each behavior is off, empty, or set to the earlier platform behavior until you change it.
Don’t enable Keypad tone for PCI or secure-input flows until secure-input suppression is in place. Real keypad tones reproduce the entered digits in the recording.
When you turn on Voice gateway turn detection, set Patience to control how long the gateway waits before it decides the caller has finished speaking.Faster settings reply sooner but may cut off callers who pause mid-sentence.
Configure speech recognition On the Speech tab, under Speech Recognition, set the following fields. The available fields vary by provider. To configure language-specific recognition, add a language profile first. A profile can contain speech synthesis, speech recognition, or both. Under Language, choose a language or enter a language tag (for example en-US), then select Add language. Configure speech synthesis Under Speech Synthesis, set the following fields. The available fields vary by provider.
  • Use the provider’s required language format. For example, ElevenLabs expects an ISO 639-1 code such as en, not en-US.
  • After you add a language, the platform creates a dedicated TTS configuration section for that language. Configure the provider, voice, model, and any provider-specific settings independently for each language.
  • Available fields vary by provider. Some providers expose additional settings such as TTS model, output format, speed, temperature, and voice tuning controls.
Configure conversation controls Set the following fields to tune barge-in, listening mode, speech timeout, and no-input behavior for the caller experience. Select one of the following Listening mode options.
If you pair Background with Barge-in off, it behaves like a slower Turn-taking mode, and Studio displays a warning.
Voice transfer and survey capabilities A read-only panel shows how many of the transfer and survey capabilities Runtime evaluates as available for the concrete voice execution boundary (the execution surface is KoreVG pipeline), for example 11 of 11 capabilities available. It lists the following capabilities, each marked Available or not. Bind the channel to an environment On the Deployment tab, under Environment Binding, use the Environment dropdown to select which deployed version handles incoming messages. Click Save Changes to apply the binding. The button is available only after you change the environment. Save the configuration Click Save Changes to apply the configuration. Each tab has its own Save Changes button, so save on each tab you edit. Confirm the selected deployment, phone number, locale, voices, and timeouts before you activate the channel.
If saving shows the warning Voice gateway synchronization is incomplete. Your configuration is saved; retry Save after the gateway recovers., the platform saved your settings but couldn’t sync them to the voice gateway. Click Save Changes again after the gateway recovers.

Test and verify the channel

Open the Testing tab to test the channel before you go live. Under Shareable softphone link, select Generate link to create a link that lets anyone with the link call this channel from a phone browser through the real phone gateway and see a live, read-only transcript. Links expire after 7 days, and you can revoke them at any time. Active links lists the links currently in use, or shows No active links for this channel. You can view active links with read access. Generating or revoking links needs permission to edit the channel. You can also use Soft Phone > Launch dialer on the Settings tab to call the agent from your browser with no DID purchase or telephony charges. Verify and secure the installation Before you go live, validate every phase of the voice pipeline.
1

Run the HTTPS preview

Grant microphone access, start and stop a call, mute and unmute, and verify that the experience fails clearly if either speech provider is unavailable.
2

Place a live phone call

Through the provisioned number, validate the greeting, STT transcript accuracy, agent response, TTS playback, barge-in, tools, transfer or hang-up, and session continuity.
3

Measure STT, LLM, TTS, and full-turn latency

Record the STT first partial and final result, the agent’s first token and completion, TTS stream open and first audio, full-turn p50 and p95, jitter, packet loss, and provider errors.
4

Exercise silence and interruption

Test silence through every no-input retry, short utterances near endpointing thresholds, speech during prompts, repeated barge-in, and rapid turns without duplicate execution.
5

Test independent phase failures

Disconnect STT, fail the agent model or a tool, reject TTS, exhaust capacity, and interrupt the network. Confirm that each phase produces the authored fallback and a clear trace.
Don’t point shared softphone registration, device call, Jambonz API, or SBC hosts at a local machine. Shared telephony endpoints stay on the environment. Local testing may change only a dedicated connection callback.
Follow these security and reliability practices:
  • Treat STT, TTS, carrier, and gateway credentials as administrator-managed secrets. Keep provider keys out of connection identifiers, browsers, agent prompts, and screenshots.
  • Set explicit phase and turn latency budgets. Avoid unnecessary transcoding and buffering, and keep prompts, tool output, and synthesis chunks small for fast first audio.
  • KoreVG processes each session sequentially through a bounded queue. Load-test rapid turns and concurrent calls to confirm that backpressure delays or rejects calls predictably.
  • Define what the caller hears for low-confidence or empty STT, model or tool failure, TTS rejection, no-input exhaustion, transfer failure, and gateway disconnect.
  • Capture scrubbed voice_stt, voice_tts, and voice_turn traces. Include call and session IDs, phase latency, interruptions, errors, provider names, cache state, and termination reasons.

Manage an existing Pipeline Voice channel

Open a channel from Deployments > Channels > Voice > Pipeline Voice. The connection has four tabs. Changing settings on the Settings, Speech, or Deployment tab updates the live channel after you click Save Changes. At the bottom of the Settings tab, the following identifiers are available. Select the copy icon to copy a value. Review recommended voice defaults When the project mixes voice and non-voice channels and the recommended voice defaults aren’t active, the Settings tab shows the banner Recommended voice defaults aren’t active. Review the voice-scoped profile. Non-voice sessions aren’t changed. Select Review voice defaults to review the profile, or Dismiss to hide the banner. For details on the profile, see Voice low-latency profile.

Troubleshooting

  • Needs phone provisioning: the connection has no phone number. Assign or buy a number under Call routing on the Settings tab before you route phone calls. A phone-routed connection can’t be activated until it has a provisioned number.
  • Voice gateway synchronization is incomplete: the platform saved your configuration but couldn’t sync it to the voice gateway. Click Save Changes again after the gateway recovers.
  • Save is rejected on a Bring Your Own SIP connection: Outbound calls is on but no SIP termination URI is set. Add a SIP termination IP address or domain name, or turn off Outbound calls.
  • Speech doesn’t work or the preview fails: confirm that both the STT and TTS providers have credentials configured, then save the connection before you preview or activate it.
  • No voice for a language: use the provider’s required language format. ElevenLabs expects an ISO 639-1 code such as en, not en-US.
  • Caller is cut off mid-sentence: raise Patience under Voice gateway turn detection, or increase Speech Timeout (ms) and Transcript Grouping Timeout (ms). For the full validation checklist, click Setup instructions on the Pipeline Voice channel page.

Pipeline vs. realtime voice

Pipeline voice (speech-to-text, agent, text-to-speech) and Realtime LLM Voice (speech-to-speech) build the agent’s turn differently underneath the same ABL. Knowing which one a channel uses matters when authoring instructions, reasoning about history after a handoff, or sizing tool results. Handoff replay on realtime channels varies by provider. OpenAI Realtime and Azure OpenAI Realtime resend the new agent’s prompt with a replay of up to 8 recent messages (4,000 characters total, 700 per message). Amazon Nova Sonic opens a new stream with recent turns in a short block (also up to 4,000 characters, 700 per message). Grok Realtime replays every earlier caller utterance. Gemini Live starts a fresh session and replays history.

Authoring voice behavior in ABL

The Studio settings on this page configure a channel. The agent’s own spoken behavior is authored in ABL:
  • Spoken-style rules and grounding guards — EXECUTION.voice.instructions and EXECUTION.voice.grounding_guards, which replace the platform’s default voice formatting rules and can refuse ungrounded tool arguments or figures. See Voice instructions and grounding guards.
  • Realtime delivery tone — EXECUTION.voice.coordinator_policy (tone, pace, brevity, interruption handling, accessibility read-backs). See Voice coordinator policy.
  • Per-response voice overrides — the VOICE: block on a RESPOND (SSML, plain text, provider, voice, speed) plus the step- and response-scoped BARGE_IN, STREAMING, and END_OF_SPEECH_TURN_TYPE overrides. See Voice configuration and Scoped barge-in.
  • Listening and turn-taking — CONVERSATION: listening: (barge-in, no-input retries, caller backchannel, on-hold phrases, barge-in continuation merging). See Voice turn-taking.
  • Low-latency defaults — new projects and voice channels get the customer_service_voice_low_latency_v1 profile automatically: tighter compaction, the fast model tier for every operation, reasoning off, and inline gather on. The Studio side of this profile — model routing, compaction budgets, and filler tuning — is covered in Voice low-latency profile.
The low-latency profile doesn’t fall back to the project’s default model. If the workspace has no model assigned to the fast tier, voice turns have no model to call.