> ## Documentation Index > Fetch the complete documentation index at: https://chatbase.co/docs/llms.txt > Use this file to discover all available pages before exploring further. # Voice > Choose how your AI agent sounds and behaves in voice conversations. Go to **Build > Voice** to configure how your AI agent sounds and behaves during voice conversations. > [!TIP] > > Want to hear it in action first? [Try the new voice experience](https://www.chatbase.co/voice). ## Voice Mode Choose how your AI agent talks with callers. There are two voice modes: * **Live voice** (recommended): One realtime model listens and speaks, so replies come almost instantly and callers can interrupt naturally. Pick a ready-made voice and go. * **Classic voice**: A separate transcriber and voice, with a short pause before each reply. It supports 100+ languages, custom or cloned voices, and fine-tuning for noisy places. ![Voice mode settings](https://mintcdn.com/chatbase/vbM66E3x4BYGeeY4/user-guides/chatbot/images/settings/voice-mode.png?fit=max\&auto=format\&n=vbM66E3x4BYGeeY4\&q=85\&s=8ba5111108c6e4f803c8a356e837a473) | | Live voice | Classic voice | | - | - | - | | Reply speed | Almost instant | Short pause before each reply | | Languages | English, Portuguese, French, German, Japanese, Hindi and Korean | 100+ | | Settings below | [Live Voice](#live-voice) | [Transcriber](#transcriber) and [AI Agent Voice](#ai-agent-voice) | ### Switching Modes Click the mode you want, then click **Switch to Live voice** or **Switch to Classic voice** to confirm. Nothing changes until you confirm, and you can click **Cancel** to keep your current mode. > [!NOTE] > > Saving puts your changes in a draft. Your customers see them once you click **Publish** in the top-right corner. A few settings go live as soon as you save. See [Drafts and versions](https://chatbase.co/docs/user-guides/chatbot/versions.md#settings-that-go-live-right-away). ## Credit Usage Voice sessions consume credits based on two components: * **AI agent voice**: Every voice session consumes **6 message credits per minute**. This is a per-minute cost regardless of the actual messages exchanged. * **AI model**: Each response from the AI agent during a voice session consumes message credits based on the model used, just like in text conversations. See the [full list of models and their message credit costs](https://chatbase.co/docs/user-guides/chatbot/build.md#ai-model). The total credit cost of a voice session is the sum of both components. ![Voice model settings](https://mintcdn.com/chatbase/P4P4lN52AUAJVwNj/user-guides/chatbot/images/settings/voice-credit-usage.png?fit=max\&auto=format\&n=P4P4lN52AUAJVwNj\&q=85\&s=bd77feeaafd85f398c75b9400ea72e08) > [accordion] Pricing Example > > Suppose your voice AI agent uses **GPT-6 Luna** (1 message credit per request). > > If a user has a **10-minute voice call** with **15 AI requests**: > > | Component | Calculation | Message Credits | > | - | - | - | > | AI agent voice | 10 min x 6 credits/min | 60 | > | AI model (GPT-6 Luna) | 15 requests x 1 credit/request | 15 | > | **Total** | | **75** | ## Voice Model Select the AI model used specifically for voice sessions. This model is separate from the one used for text-based conversations and only applies when users interact with your AI agent through voice. You can also adjust the **Temperature** to control how creative or reserved the AI agent's responses are during voice conversations. ![Voice model settings](https://mintcdn.com/chatbase/P4P4lN52AUAJVwNj/user-guides/chatbot/images/settings/voice-model-and-temperature.png?fit=max\&auto=format\&n=P4P4lN52AUAJVwNj\&q=85\&s=31210835051fdd5ddf5146132084643b) ## Live Voice When your AI agent uses Live voice, choose its voice and describe how it should sound. ![Live voice settings](https://mintcdn.com/chatbase/vbM66E3x4BYGeeY4/user-guides/chatbot/images/settings/voice-live-voice.png?fit=max\&auto=format\&n=vbM66E3x4BYGeeY4\&q=85\&s=3f4e008932e539cecb04215fbe2b16a8) * **Voice**: Click the voice selector to browse the Live voices, preview them, and pick one. * **Speaking style**: Describe how the voice should sound, such as its tone, pace, formality and emotion. What your AI agent says still comes from its instructions. Up to 500 characters. For example: ``` Speak warmly and naturally, at an unhurried pace. Sound friendly but professional, and use everyday words. If a caller sounds frustrated, stay calm and reassuring. ``` Tips for a good speaking style: * Write it in the language your AI agent speaks. * Describe the sound, not the content. Put rules about what to say in your instructions. * Be specific about pace, for example "speak slowly and clearly" rather than "speak well". If you leave it empty, Live voice uses a natural, friendly default. Make sure to click **Save** on the card to apply your changes. ## Transcriber > [!NOTE] > > The transcriber only applies to Classic voice. Choose the model used to transcribe user speech during voice conversations. ![Voice transcriber settings](https://mintcdn.com/chatbase/3sPesZY8zXVkuyp6/user-guides/chatbot/images/settings/voice-transcriber.png?fit=max\&auto=format\&n=3sPesZY8zXVkuyp6\&q=85\&s=3d985a9ecad66f89b6276ddc5882e020) * **Language**: The language used to transcribe user speech. Select **Multilingual (auto-detect)** to let the transcriber detect the language automatically, or pick a specific language to improve accuracy. * **Model**: The speech-to-text model that converts user audio into text. Each model supports a different range of languages. * **Silence Threshold**: Seconds of silence before the AI agent considers the user has finished speaking. Lower values switch turns faster but may cut off pauses mid-thought. * **Noise Sensitivity**: Controls how loud audio must be to count as speech. Lower values pick up quieter speech but are more sensitive to background noise. Higher values filter out noise but may miss soft-spoken users. > [!NOTE] > > **Multiple transcription languages:** When using the default **Soniox v5** model, you can select up to **8 languages**. Limiting the set of possible languages helps the transcriber focus on the selected languages, improving transcription accuracy. ## AI Agent Voice > [!NOTE] > > These settings only apply to Classic voice. For Live voice, see [Live Voice](#live-voice). Choose a voice for your AI agent and fine-tune how it sounds during voice conversations. ![Voice AI agent voice settings](https://mintcdn.com/chatbase/3sPesZY8zXVkuyp6/user-guides/chatbot/images/settings/voice-agent-voice.png?fit=max\&auto=format\&n=3sPesZY8zXVkuyp6\&q=85\&s=4906225649f908afe1cb89db82c7e362) Click on the voice selector to open the voice picker, then fine-tune how the selected voice sounds. The available controls depend on the voice you picked, some voices expose sliders such as speed, stability, and similarity, while others let you describe the desired delivery. A few examples: * **Speed**: Controls how fast or slow the AI agent speaks. Adjust the slider between slower and faster to match your preferred pace. * **Stability**: Determines how stable the voice is. Lower values introduce a broader emotional range, while higher values produce a more consistent, monotone delivery with limited emotion. * **Similarity**: Determines how closely the AI adheres to the original voice when replicating it. Higher values result in a closer match to the selected voice. * **Voice instructions**: For voices that support it, enter a short prompt describing the tone, style, or dialect you want the AI agent to use when speaking. Make sure to click **Save** on the voice card to apply your changes. > [!NOTE] > > Saving puts your changes in a draft. Your customers see them once you click **Publish** in the top-right corner. A few settings go live as soon as you save. See [Drafts and versions](https://chatbase.co/docs/user-guides/chatbot/versions.md#settings-that-go-live-right-away). ### Selecting a voice The voice picker has two tabs: **All voices** for browsing pre-recorded voices, and **Custom voice** for bringing a custom voice from a supported provider. ![Voice selection modal](https://mintcdn.com/chatbase/3sPesZY8zXVkuyp6/user-guides/chatbot/images/settings/voice-select-voice.png?fit=max\&auto=format\&n=3sPesZY8zXVkuyp6\&q=85\&s=81a4f846549c9e448efa3636fed24ac2) Under **All voices**, you can search by name and filter voices by language, accent, age, and gender. Each row shows the voice's supported languages and a **Preview language** dropdown, pick a language and click the play button to hear a sample of the voice speaking it. Select a voice and click **Confirm voice** to apply it to your AI agent. ![Custom voice selection modal](https://mintcdn.com/chatbase/3sPesZY8zXVkuyp6/user-guides/chatbot/images/settings/voice-select-custom-voice.png?fit=max\&auto=format\&n=3sPesZY8zXVkuyp6\&q=85\&s=d317fbc2ec67fddb3887f5df55dabe36) Under **Custom voice**, you can add a custom public voice that isn't in the pre-recorded list by referencing it directly from your provider: * **Provider**: The voice provider that hosts the custom voice (for example, ElevenLabs). * **Voice ID**: The identifier of the voice in the provider's account. Paste it in and click **Add voice** to make it available for selection, then click **Confirm voice** to apply it to your AI agent. > [!NOTE] > > Custom voices only work with **publicly available shared voices** from your provider. Private voices in your provider account are not supported. ### Using Dialects and Regional Accents Some voices support custom dialects and accents through Voice Instructions. Look for voices marked with the ✨ sparkle icon. These voices can adapt their accent, dialect, pacing, and speaking style based on the instructions you provide. ![Voices Sparkle](https://mintcdn.com/chatbase/8Z0YT4pV3LwKN868/user-guides/chatbot/images/settings/voices-sparkle.png?fit=max\&auto=format\&n=8Z0YT4pV3LwKN868\&q=85\&s=fdf5ece517a0a15062cb75bada862a85) After selecting the desired voice, in the Instructions field, describe how you want the voice to sound. **Recommended Settings** * Transcriber: Soniox * Voice LLM: GPT-6 Luna or Gemini 3.8 Flash ### Pronunciations > [!WARNING] > > **Pronunciations only work with ElevenLabs voices.** If your agent uses a voice from any other provider, you will not see this option. Use this to fix words your agent says wrong, like brand names, product names, or acronyms. ![Voice pronunciations settings](https://mintcdn.com/chatbase/8Z0YT4pV3LwKN868/user-guides/chatbot/images/settings/voice-pronunciations.png?fit=max\&auto=format\&n=8Z0YT4pV3LwKN868\&q=85\&s=3636bbf4b91e39dc11cc60e5af26ead0) Each row has two fields: * **Word or phrase**: The word as it is written, for example `SQL`. * **Spoken as**: How the agent should say it, spelled the way it sounds, for example `sequel`. Click **Add pronunciation** to add a row, or the trash icon to remove one. Click **Save** when you are done. Every time the word comes up in a voice conversation, the agent says it the way you spelled it. You can add up to 200 words, and each word can only have one pronunciation. ## Messages Configure the initial greeting and error messages for your voice AI agent. ![Voice messages settings](https://mintcdn.com/chatbase/P4P4lN52AUAJVwNj/user-guides/chatbot/images/settings/voice-messages-section.png?fit=max\&auto=format\&n=P4P4lN52AUAJVwNj\&q=85\&s=c4462dfedbe7df278224fe83caf4f58f) * **Initial message**: The first message the AI agent speaks when a voice session begins. If left empty, the voice AI agent will wait for the user to speak first. * **Error message**: The message played if the voice session encounters an error. If left empty, the default error message "The voice session encountered an error. Please try again." will be used. ## Session Configurations Control how the AI agent handles interruptions, silence, and text input during active voice calls. ![Voice session configurations settings](https://mintcdn.com/chatbase/Y3la_0-l9aJURKJI/user-guides/chatbot/images/settings/voice-session-configs.png?fit=max\&auto=format\&n=Y3la_0-l9aJURKJI\&q=85\&s=2f95b152d10272e75ac5809abea38050) * **Allow interruptions**: When enabled, users can speak over the AI agent to interject mid-response. Disable this if you'd prefer the AI agent to always finish speaking before listening for the user's next turn. Defaults to **enabled**. * **End conversation after silence**: The number of seconds of inactivity after which the call is automatically ended. Useful for closing out abandoned sessions and avoiding unnecessary credit usage. Defaults to **300 seconds**. * **Allow text input during voice calls**: When enabled, users can type messages mid-call alongside speaking. Helpful for entering details that are easier to type than to say, such as names, emails, or URLs. Defaults to **disabled**. ## Recordings Enable recording of voice sessions for quality review and training purposes. You can also toggle **Retain voice recordings forever** to keep recordings indefinitely, or set a custom retention period. ![Voice recordings settings](https://mintcdn.com/chatbase/P4P4lN52AUAJVwNj/user-guides/chatbot/images/settings/voice-recordings-section.png?fit=max\&auto=format\&n=P4P4lN52AUAJVwNj\&q=85\&s=9581650f1d156a10a0a21d1e414bc6f3) ## Limits Control session volume and duration to manage costs and performance. ![Voice limits settings](https://mintcdn.com/chatbase/P4P4lN52AUAJVwNj/user-guides/chatbot/images/settings/voice-limits-section.png?fit=max\&auto=format\&n=P4P4lN52AUAJVwNj\&q=85\&s=21cba74d863f24a92ebcd72bfc6b6727) * **Max no. of concurrent voice sessions**: The maximum number of voice sessions that can run at the same time. If left empty, your plan's default limit will apply. * **Max call duration**: The maximum length of a single voice call in minutes. The default is 15 minutes. * **Allowed no. of daily calls**: The total number of voice calls allowed each day. If left empty, your plan's default limit will apply. > [!NOTE] > > Your subscription plan enforces account-level limits for daily sessions and concurrent sessions across all your AI agents. You can set agent-specific limits here to further restrict usage for individual AI agents. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.