> ## Documentation Index
> Fetch the complete documentation index at: https://docs.monobot.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Speech and Voice

This page describes available actions used to control call behavior, voice interaction within your bot.

Actions are building blocks of your flow. Each action performs a specific function — such as capturing user input, playing messages or transferring calls.

Configure actions by defining required parameters and optional settings based on your use case.

<Frame>
  <img className="block" src="https://mintcdn.com/monobot-0e0ffe36/90B0L5Hnl7m86xGy/images/speech_voice_image/speech_voice.png?fit=max&auto=format&n=90B0L5Hnl7m86xGy&q=85&s=62a5569730169bb4b9092742f948b9d0" alt="Speech Voice" width="3027" height="1619" data-path="images/speech_voice_image/speech_voice.png" />
</Frame>

<Accordion title="DTMF">
  Captures keypad input from the caller during a call.

  ### Required data

  * **DTMF Value**\
    Type: `string`\
    Variable used to store the pressed key.

    *Example:* `%dtmf%`

  * **Output**\
    Type: `string`\
    Message returned after input is received.

    *Example:* `DTMF signal %dtmf% was received.`

  ### Notes

  * Used for IVR menus (Press 1 / Press 2)
  * Can be used in conditions to route calls
</Accordion>

<Accordion title="Hang Up">
  Terminates the active call.

  ***

  ### Optional data

  * **Answer**\
    Type: `string`\
    Message played before the call ends.

    *Example:* `Thank you for calling. Goodbye.`

  ***

  ### Notes

  * Immediately ends the call after execution
</Accordion>

<Accordion title="Language Switcher">
  Switches language and voice processing during the call.

  ***

  ### Required data

  * **Language**\
    Type: `string`\
    Language for recognition and responses.

    *Example:* `English`

  ***

  ### Optional data

  * **STT Processor**\
    Type: `string`\
    Speech-to-text engine.

    *Example:* `Flux General`

  * **VAD Model**\
    Type: `string`\
    Detects when user starts/stops speaking.

    *Example:* `None`

  * **EOT Model**\
    Type: `string`\
    Detects end of user speech.

    *Example:* `None`

  * **TTS Processor**\
    Type: `string`\
    Text-to-speech engine.

    *Example:* `Deepgram Aura 2`

  * **TTS Voice ID**\
    Type: `string`\
    Voice configuration.

    *Example:* `aura-2-andromeda-en`

  ***

  ### Notes

  * Affects both input (STT) and output (TTS)
  * Use for multilingual flows
</Accordion>

<Accordion title="Redirect Call">
  Transfers the call to another destination.

  ***

  ### Required data

  * **Transfer To**\
    Type: `string`\
    Phone number, extension, or SIP URI.

    *Examples:*\
    `+14155551234`\
    `222222`\
    `sip:222222@example.com`

  ***

  ### Optional data

  * **Is SIP**\
    Type: `boolean`\
    Enable if destination is SIP.

  * **Introduction**\
    Type: `string`\
    Message before transfer.

  ***

  ### Notes

  * Works only if a phone number is attached to the bot
</Accordion>

<Accordion title="Say">
  Plays a voice message to the caller.

  ***

  ### Required data

  * **Announcement Text**\
    Type: `string`\
    Text to be spoken.

    *Example:* `Please hold while I connect you.`

  ***

  ### Optional data

  * **Terminate LLM answer**\
    Type: `boolean`\
    Stops current AI response.

  * **Incognito call**\
    Type: `boolean`\
    Prevents storing/logging message.

  ***

  ### Notes

  * Used for system messages and prompts
</Accordion>