Skip to main content
Famulor selects the models used to understand and answer calls. You choose voices, languages and speaking controls. The AI Safety & Control add-on also enables compatible model choices for your workspace. The model picker and your assistant’s selected model names are visible only when AI Safety & Control is available in the active workspace. This rule applies to the dashboard, REST API and MCP; Whitelabel access alone does not grant model access. Without this add-on, the models selected automatically for your assistant are not shown. Voice selection and language controls remain available according to your plan.

​
Model catalog

The assistant editor shows a curated catalog of models available to your workspace: Availability depends on your workspace and selected engine mode. When AI Safety & Control is included, you can choose models separately for pipeline, realtime, and half-cascade assistants. Otherwise, the assistant uses the recommended automatic selection. See AI regions & models for a reference of available models, EU/US endpoint availability, automatic SIP routing and infrastructure locations. New language-model choices for Pipeline include GPT-6 Sol, DeepSeek V4 Pro, Kimi K2.6, GPT-OSS 120B, Grok 4.3, Grok 4.5 and Grok 4.6. GPT-Live Transcribe and English-only Inworld STT 1 provide recognition for Pipeline. Gradium TTS Beta provides speech output in English, French, German, Spanish and Portuguese for Pipeline and Half-Cascade. Only enabled choices compatible with your workspace region are shown. GPT-Live Transcribe is separate from the Full Duplex conversation engine. Some language models use fixed sampling settings, so changing Temperature has no effect on them. When enabled for your workspace, Grok Voice 3.0 uses Realtime → Standard with its own public voices, including Eve. It does not use a separate TTS voice. Half-Cascade and Full Duplex are not supported for this model. It has no dedicated EU endpoint and is unavailable in EU workspaces. Use Standard Realtime in the voice catalog to find its compatible voices. API and MCP users can browse them with mode=realtime and realtime_variant=standard; stored previews use the returned authenticated preview link. With model-selection access, choose a compatible model in the same update when switching to a voice from another model. CLI: famulor list-voices --mode realtime --realtime-variant standard.

​
Grok Voice 3.0

Grok Choose Grok Voice 3.0 under Realtime → Standard and select one of its 28 public voices. The same conversation model receives speech and produces the spoken reply; you do not select a separate TTS voice for the conversation. Start with Eve, listen to previews, save, and test a call with your own tools. Native speech-to-speech can reduce the steps between listening and replying. Actual response time depends on the conversation, connection and tool execution; no fixed end-to-end latency is guaranteed.

​
Primary speech recognition model

With AI Safety & Control, open Assistant settings → General → Advanced. In Pipeline mode, Speech recognition model (STT) appears below the language model. Choose an available provider and model, or return to Platform default. The list follows your workspace’s AI region. The selection is hidden and inactive for Realtime, Full Duplex, Half-Cascade and Translate. Switching engines preserves your Pipeline preference. Only the primary recognition model changes. The platform continues to manage the fallback models and their order. If the add-on is removed or a saved model is no longer available in the workspace region, subsequent calls use the central default. REST: list compatible choices with GET /api/v1/models?type=stt&mode=pipeline, then update the assistant with PUT /api/v1/assistants/{id}. MCP: use get_models and update_assistant. Set the recognition provider and model together; set both to null to return to the default.

​
Temperature

In Pipeline mode, Temperature controls how closely the language model sticks to your prompt versus varying its wording. The slider runs from 0 — deterministic and on-script — upward to more creative, improvised replies. 0.5–0.8 is a typical range for phone conversations. Treat it as final fine-tuning, not a substitute for a well-written prompt:
  1. Finish the prompt first — role, goals, boundaries, tone.
  2. Start low.
  3. Raise it in small steps, only once the baseline sounds good.
  4. Test and compare after each change.
Raise it when replies sound stiff or formulaic and the use case tolerates some improvisation. Keep it low when consistency, compliance, or precise wording matter more than natural variation.

​
Voice library

Use the voice picker to filter by language and voice characteristics, then play a sample before saving. The library includes Deepgram Aura-2, Rime Coda and Gradium TTS, with playable samples and provider filters sorted A–Z. Gradium offers 71 flagship voices across five languages. Available voices depend on your workspace’s AI region and selected languages.

​
Gemini 3.8 text-to-speech

Gemini 3.8 Flash-Lite TTS and Gemini 3.8 Flash TTS are speech-generation models for Pipeline and Half-Cascade assistants. Choose a ready-made Gemini voice and play its preview in the voice library. Flash-Lite TTS is the default for Google voices; choosing a specific compatible model requires AI Safety & Control. Both support Dynamic emotions (Expressive Mode). Delivery can adapt to the conversation; the effect of individual cues depends on the selected model, voice and text. They are available in Global and US workspaces through unrestricted Global routing. Selecting US does not restrict these two models to US processing. They are unavailable in EU workspaces. They do not replace speech recognition or a realtime engine.
Voice picker with Saved and Explore tabs, language and accent filters

Assistant editor → Change voice → Explore: filter the library by language, accent and voice characteristics. Saved contains voices you have kept.

The public REST endpoint GET /api/v1/voices and MCP tool get_voices return provider-neutral local selectors. Upstream voice IDs and supplier names are not exposed; pass the returned id directly as tts_voice. For the same voice list, use famulor list-voices --mode tts in the CLI or client.voices.list({ mode: "tts" }) in the TypeScript SDK. To list available speech-generation models, use GET /api/v1/models?type=tts&mode=pipeline, MCP get_models, famulor list-models --type tts --mode pipeline, or client.api.listModels({ query: { type: "tts", mode: "pipeline" } }). The same workspace region and model-access rules apply.
  • Assistant voice — the default voice for the assistant.
  • Voice per flow agent — give individual flow agents distinct voices.
  • Voice per language — change the voice together with automatic language switching.
  • Cloned voices — when included in your plan, add a private custom voice and use it like a library voice.
When recording a sample to clone, use clear, high-quality audio with steady, natural delivery and no background noise — and only clone a voice you have permission to use (see Voice cloning consent). For API cloning, check sample_constraints from GET /api/v1/voices/clone/capability for how many samples are currently accepted (max_files, up to 3), the maximum size per sample (max_bytes_per_file), and the maximum sample length (max_duration_seconds, or null for no limit), prepare that many direct uploads with POST /api/v1/voices/clone/uploads, upload the samples, then submit the job with POST /api/v1/voices/clone. MCP offers the equivalent get_voice_clone_capability, create_voice_clone_upload, and submit_voice_clone tools. The dashboard, REST, and MCP paths share the same Clone your own voice entitlement, consent check, creation pricing, and tenant isolation.

​
Speaking style

The available controls adapt to the selected model. Depending on the voice, you may see:
  • speaking rate and output volume;
  • stability, similarity, or expressiveness;
  • free-text style instructions;
  • a pronunciation dictionary for names, abbreviations, and specialist terms;
  • filters that prevent markdown and emoji from being read aloud.
Only compatible controls are shown and applied.
Pick the voice first, then tune one control at a time — stability, similarity, or rate — and listen to a realistic phrase from your actual call flow before changing the next one.
When the Dynamic emotions setting (Expressive Mode) is available for the selected voice, the assistant can adapt emotion, pacing, emphasis, pauses, and supported non-verbal sounds to the conversation. It applies to pipeline and half-cascade assistants; realtime speech models handle vocal expression natively. Availability depends on the selected voice, and unsupported delivery cues are not applied. Transcripts remain clean and do not include delivery markup.

​
Speech recognition glossary

Open Assistant settings → Voice → Speech recognition to add customer, product, and proper names that should be transcribed accurately. The glossary affects the transcript; use the pronunciation dictionary separately when a term also needs a special spoken form. The glossary is shown only when the effective recognition model supports saved terms in your workspace. The editor displays that model’s term and length limits, and an additional-usage notice where applicable. Changing models preserves your saved terms, but unsupported models do not use them. Some models also limit the total vocabulary size. In that case, only the first terms fitting the limit are used; the full saved list remains available when you switch models. Saved glossary terms and automatic detection have separate requirements. The glossary can be available even when automatic detection is not. Adding saved terms does not enable automatic detection. If Automatic term detection (Beta) is available and enabled, the assistant can recognise additional terms for the current call. This may add AI usage. Detected terms are not added to your saved glossary.

​
Fallback chains

Fallback chains is a single switch, not a set of choices you configure. Turn it on, and if the primary speech-recognition, language-model, or voice service times out, errors, or drops mid-call, the assistant automatically moves to a compatible backup so the conversation keeps going instead of dropping. Famulor selects and maintains which backup each chain falls through to — there’s no picker for the backup models themselves. For Pipeline voice and text chat, a working language-model backup stays selected until it also fails. This avoids repeatedly switching back to an unstable primary. New conversations and assistant handoffs start with that assistant’s selected primary again. This policy does not replace native Realtime, Half-Cascade or Full Duplex models, and it does not change speech-recognition or voice fallbacks.
Fallback chains ship with the AI Safety & Control add-on; check Settings → Plan for availability. Failing over doesn’t add a separate charge on its own — a call that fails over still bills at the normal call-minute rate.
Test important assistants after changing models, voices, or languages. A short simulation is usually enough to catch pronunciation and timing differences.

​
Saved announcements

Prepare and preview each fixed announcement using compact icons inside its text field:
  • Greeting on the assistant’s main page: the greeting and any enabled static message after silence.
  • Assistant settings → Privacy → Consent: the consent message.
  • The end-call tool editor: that action’s farewell message.
When secondary languages are configured, the farewell text stays editable and guides a message generated in the caller’s current supported language. These farewells do not use saved audio; preparation skips them. Greetings and consent keep their existing preparation behavior. Without secondary languages, farewells keep their existing delivery behavior, including supported saved audio. Use the Play, Stop and Generate audio icons inside the relevant field. A tooltip shows the preparation status and recording duration when available. Generation uses the saved settings, so save draft changes first. Saving requests preparation in the background. A compact animated dialog confirms your save, and you can continue editing immediately afterward. Preparation details remain beside each message, without interrupting your work. Matching audio is reused. Changing a message, voice, language or relevant speaking settings requires a matching recording before that audio can be used again. The preparation keeps the selected engine’s supported voice. Full Duplex uses its native voice; Fallback settings applies only if the conversation switches to Pipeline. Uploaded greetings retain their recorded voice. Caller-specific variables, dynamic messages, unsupported modes and unsuccessful preparation keep their existing call behavior. A caller-first greeting still waits for the caller; eligible fixed messages after silence are prepared separately. Preparation does not prove that a consent message was heard or accepted. Consent is still checked during each call, including complete playback where required. REST supports GET /api/v1/assistants/{id}/announcements, POST on the same path, and the authenticated audio link returned for a ready item. MCP and Milian provide get_assistant_announcements and prepare_assistant_announcements. Audio previews are private to the workspace.

​
Voice creation and credits

Voice cloning is a plan feature. Open Clone a voice directly; the creation button shows Create · free while your allowance remains, then Create + X credits. The charge is made once for a successful creation. Stored voices have no plan capacity limit or recurring storage charge. Send expected_credits with POST /api/v1/voices/clone or submit_voice_clone, using the displayed pricing.next_creation_credits from the capability response. Set it to 0 to accept only a free creation. If the price increases or the last free creation is used elsewhere, refresh the price and confirm again. Retry the same job ID after an interrupted request; a successful job is charged only once. The legacy clone_voice tool also requires expected_credits.

​
Backup voices

In Settings → Conversation → Fallbacks, enable fallback chains and choose a voice for each available step. Preview voices in the same library as your main voice. The diagram shows the order: primary voice, then up to three backups. Use default clears a choice. Save the assistant to apply changes. Requires Fallbacks & Guardrails. The platform manages backup models and their order; the library shows compatible public voices. Private cloned voices are not available as these backups. Pipeline and Half-Cascade use these speech backups. In Full Duplex they apply after switching to backup Pipeline speech. Turning fallbacks off keeps your saved voice choices without activating them. REST: GET /api/v1/assistants/{id}/fallback-voices, PATCH /api/v1/assistants/{id}/fallback-voices, GET /api/v1/voices?fallback_position=1. MCP: get_assistant_fallback_voices, update_assistant_fallback_voices, get_voices (fallback_position).