Actions39
- Account Actions
- Auth Actions
- Bluetooth Actions
- Conversation Actions
- Device Actions
- Interaction Actions
- List Actions
- Notification Actions
- Routine Actions
- Smarthome Actions
Interaction → Speak
AI-generatedSummary
Make an Alexa device speak text or SSML markup.
Inputs
- device (required) — Select an Echo device from the list or provide its serial number via an expression.
- text (required) — Text to speak. Visible when Speak Type is set to 'Regular' or 'Announcement'.
- ssmlContent (required) — SSML markup to send. Visible when Speak Type is set to 'SSML'. Must be wrapped in <speak> tags. Supported tags include <break time="500ms"/>, <emphasis>, <prosody rate="slow">, and <say-as interpret-as="characters">.
- speakType — Type of speech: 'regular' (standard text-to-speech), 'ssml' (structured markup), or 'announcement' (multi-device broadcast). Defaults to 'regular'.
Output shape
A single record with a 'success' property indicating the operation status.
If the device is part of a multi-room group, the command is executed on all group members.
Examples
Example 1: Speak a simple greeting on a single device.
Device: Kitchen Echo; Text: 'Good morning'; Speak Type: Regular
Example 2: Send an announcement to all devices in a multi-room group.
Device: Living Room Echo; Text: 'Meeting starting in 5 minutes'; Speak Type: Announcement
Example 3: Speak with specific timing using SSML.
Device: Bedroom Echo; SSML Content: '