Skip to main content

Text-to-Speech in Alerts

Text-to-Speech (TTS) reads alert messages aloud on your stream. When a viewer cheers, subscribes, or sends a Super Cheer with a message, TTS can voice that message so you and your chat can hear it.

Enabling TTS

TTS is configured per stream trigger in the trigger's global settings:

SettingDescription
EnableTurn TTS on or off for this trigger
Default VoiceThe voice used when no specific voice is selected
VoicesList of available voices for randomization or user selection
Randomized VoiceRandomly pick a voice from the list for each alert
Allow User ChoiceLet viewers select their preferred voice (Super Cheers)
Read EventRead the event type (e.g., "new follower from...")
Read MessageRead the viewer's message content

TTS Timing

Control when TTS plays relative to the alert:

SettingDescription
TTS DelaySeconds to wait before TTS starts (per event type)
TTS Start After AudioWait for the alert's sound effect to finish before TTS begins

These settings are configured per event type (follows, cheers, subs, etc.) in the Alert Box widget.

Available Providers

Streamerly supports multiple TTS providers, each with different voice qualities and options:

ProviderEngine TypesNotes
ElevenLabsAIHighest quality, most natural voices. Supports voice cloning
Microsoft AzureStandard, NeuralWide language and voice selection
Amazon PollyStandard, NeuralReliable with many voice options
DECTalkDECTalkRetro-style synthesized voices (Paul, Betty, Harry, Frank, Dennis, Kit)

For a detailed comparison and configuration guide, see the TTS documentation.

See Also