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

# Voice Capabilities

Chat includes voice input and voice output capabilities that let you speak your messages and have AI responses read back to you.

> **Voice Input**
>
> Dictate messages into the chat input field. Transcribed text is inserted at the cursor position in real time.

> **Text-to-Speech**
>
> Have AI responses read aloud. Pause and resume playback from a mini-player pill in the input area.

> **Speaking Mode**
>
> Hands-free voice conversation loop. ELITEA records, sends, speaks the response, and listens again — automatically.

### Voice Input

**Voice Input** lets you dictate messages directly into the chat input field using a microphone. The transcribed text is inserted at the **current cursor position**, so you can combine typed and spoken content in the same message — position your cursor anywhere in the text before starting to record.

**How to use Voice Input:**

1. (Optional) Click inside the input field and place the cursor where you want the dictation inserted.
2. Click the **microphone** icon in the message input toolbar (tooltip: *Start voice input*) to start recording.
3. The mic icon turns **teal** and a **Stop (■)** button appears beside it in a pill-shaped container — this indicates active recording.
4. Speak your message. A live transcript appears in the input field in real time. Interim (in-progress) words are shown immediately and replaced by the final transcript as recognition completes.
5. Click the **Stop (■)** button (tooltip: *Stop dictation*) to finish recording. The final transcript is committed at the cursor position and focus returns to the input field.
6. Review the transcribed text, then press **Send** to send your message.

> [!NOTE]
>
> The **Send** button is disabled while recording is active. Stop recording before sending.

![Voice Input](https://mintcdn.com/epam-a74ef051/t-vUROB65VJXdcH6/img/menus/chat/voice-input.gif?s=88461191679ec9a6bd9a8669d66ab788)

> [!NOTE]
>
> **ASR model selection.** ELITEA selects the best available speech recognition source in the following priority order:
>
> 1. **Server-side Realtime/Streaming ASR model** (default model first, if configured in [Settings → AI Configuration → Speech Recognition](https://docs.elitea.ai/menus/settings/ai-configuration.md#speech-recognition-asr))
> 2. **Server-side Whisper ASR model** (default model first — batch transcription with slightly higher latency)
> 3. **Browser Web Speech API** (built-in browser fallback, no server configuration required)
> 4. **Microphone icon hidden** — if none of the above is available, the button is not shown
>
> If an administrator has temporarily disabled voice features, the microphone icon is visible but grayed out with the tooltip *"Temporarily disabled by admin"*.

### Text-to-Speech

**Text-to-Speech** reads AI responses aloud. A **Read out** button (megaphone icon) appears in the action bar below each AI message when the message contains speakable text.

**How to use Text-to-Speech:**

1. Click the **Read out** (megaphone) button below an AI message to start playback.
2. The text is read aloud and a **playback pill** appears in the input area containing:
   * A **megaphone** icon
   * A **Stop (■)** button (tooltip: *Stop speaking*) — click to stop playback
   * A **Voice settings (⚙)** gear icon (tooltip: *Voice settings*) — opens the Voice Settings dialog. Disabled while playback is active.
3. To restart playback after stopping, click the **Play (▶)** button that replaces the Stop button in the pill.

   ![Text-to-Speech](https://mintcdn.com/epam-a74ef051/t-vUROB65VJXdcH6/img/menus/chat/text-to-speech.gif?s=427ef51ff7df9161e69af256f50f4ef8)

> [!NOTE]
>
> When a TTS model is configured in [**Settings → AI Configuration → Text to Speech (TTS)**](https://docs.elitea.ai/menus/settings/ai-configuration.md#text-to-speech-tts), ELITEA uses that model for audio generation. If no TTS model is configured, playback uses the browser's built-in SpeechSynthesis API as a fallback.

**Voice Settings:**

Click the **⚙** gear icon in the playback pill (only available when playback is stopped) to open the **Voice settings** dialog. Settings are saved to your browser and persist across sessions.

| Setting | Description |
| - | - |
| **Voice** | Select the voice to use for playback. When a server-side TTS model is configured, the list shows that model's available voices. Without a server TTS model, the list shows your browser's built-in speech synthesis voices (non-local voices are marked *"online"*). |
| **Speed** | Controls playback rate. Range: **0.5×** (slower) to **2×** (faster). Default: **1×**. |
| **Volume** | Controls playback volume. Range: **Mute (0%)** to **100%**. Default: **100%**. |
| **Preview Voice** | Plays a short sample sentence with the currently selected voice and settings. Only available when playback is not active. |

Click **Apply** to save, or **Cancel** to discard changes.

![Text-to-Speech](https://mintcdn.com/epam-a74ef051/t-vUROB65VJXdcH6/img/menus/chat/voice-settings.gif?s=45354de5a28fbce8f747f69620d8a13a)

> [!TIP]
>
> You can also configure voice preferences globally under **User Settings → Profile → Voice Personalization**. These settings apply as the default for all new TTS playbacks across all conversations.

### Speaking Mode

> [!WARNING]
>
> Speaking Mode is only available **after the conversation has been created** — meaning at least one message must have been sent first. The voicewave icon does not appear in a brand-new, unsent conversation. Send your first message to initialize the conversation, then use Speaking Mode for subsequent turns.

**Speaking Mode** is a continuous, hands-free voice conversation loop. Once activated, ELITEA automatically records your speech, sends your message after a pause, plays back the AI's response, and then starts listening again — without any manual interaction between turns.

**How to activate Speaking Mode:**

1. Leave the message input field empty.
2. Click the **voicewave** icon (shown in place of the Send button when the input is empty) to enter Speaking Mode.
3. A voicewave pill with an **Exit** (✕) button replaces the send button, indicating Speaking Mode is active.
4. Speak your message. After **3 seconds of silence**, your message is automatically sent.
5. The AI response streams in and is read aloud via Text-to-Speech.
6. Once the response finishes playing, recording starts again automatically for the next turn.
7. Click the **✕** button in the voicewave pill to exit Speaking Mode at any time.

![Speaking Mode](https://mintcdn.com/epam-a74ef051/t-vUROB65VJXdcH6/img/menus/chat/speaking-mode.gif?s=9b45d821761021ca20c7b1d1174585bb)

> [!NOTE]
>
> Speaking Mode requires a voice recognition source (server ASR model or browser Web Speech API) for input and uses Text-to-Speech for output. If you start manual voice input while Speaking Mode is active, Speaking Mode is automatically deactivated.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.