google/gemini-3.8-flash-lite-tts

OfficialView on Replicate →

Google's fast, low-cost text-to-speech model with line-by-line delivery direction, vocal bursts, and two-speaker dialogue in 100+ languages

Capabilities

No capability data available

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`text`*	string	The exact words to speak. Add momentary sounds and pauses inline with angle-bracket tags, for example "Wait... <short pause> did you hear that? <gasp>". Supported tags include <laugh>, <sigh>, <gasp>, <cough>, <breath>, <whispering>, <short pause> and <long pause>. Spell tricky names in IPA between slashes, like /niːv/. In two-speaker mode, write one line per turn as "Name: what they say" (or "Name (style): what they say" to direct that line), and wrap quick reactions from the other speaker in pipes, like "So the launch is Thursday \|oh hmm\| are we ready?".	`—`	—
`output_format`	string	Audio file format.	`"wav"`	wavmp3
`second_voice`	string	Voice for the second speaker. Set this to turn on two-speaker mode, where each line of text must look like "Name: what they say" with exactly two different names.	`"none"`	noneAchernarAchirdAlgenibAlgiebaAlnilamAoedeAutonoeCallirrhoeCharonDespinaEnceladusErinomeFenrirGacruxIapetusKoreLaomedeiaLedaOrusPulcherrimaPuckRasalgethiSadachbiaSadaltagerSchedarSulafatUmbrielVindemiatrixZephyrZubenelgenubi
`style`	string	Optional delivery direction for every line, such as "warm and enthusiastic", "whispered urgently" or "speaking slowly". Leave empty for natural delivery, which usually works best. In two-speaker mode, a "Name (style):" prefix overrides this for that line.	`""`	—
`voice`	string	Voice for the narrator, or for the first speaker in two-speaker mode.	`"Kore"`	AchernarAchirdAlgenibAlgiebaAlnilamAoedeAutonoeCallirrhoeCharonDespinaEnceladusErinomeFenrirGacruxIapetusKoreLaomedeiaLedaOrusPulcherrimaPuckRasalgethiSadachbiaSadaltagerSchedarSulafatUmbrielVindemiatrixZephyrZubenelgenubi

textrequiredstring

The exact words to speak. Add momentary sounds and pauses inline with angle-bracket tags, for example "Wait... <short pause> did you hear that? <gasp>". Supported tags include <laugh>, <sigh>, <gasp>, <cough>, <breath>, <whispering>, <short pause> and <long pause>. Spell tricky names in IPA between slashes, like /niːv/. In two-speaker mode, write one line per turn as "Name: what they say" (or "Name (style): what they say" to direct that line), and wrap quick reactions from the other speaker in pipes, like "So the launch is Thursday |oh hmm| are we ready?".

output_formatstring

Audio file format.

Default: "wav"

wavmp3

second_voicestring

Voice for the second speaker. Set this to turn on two-speaker mode, where each line of text must look like "Name: what they say" with exactly two different names.

Default: "none"

noneAchernarAchirdAlgenibAlgiebaAlnilamAoedeAutonoeCallirrhoeCharonDespinaEnceladusErinomeFenrirGacruxIapetusKoreLaomedeiaLedaOrusPulcherrimaPuckRasalgethiSadachbiaSadaltagerSchedarSulafatUmbrielVindemiatrixZephyrZubenelgenubi

stylestring

Optional delivery direction for every line, such as "warm and enthusiastic", "whispered urgently" or "speaking slowly". Leave empty for natural delivery, which usually works best. In two-speaker mode, a "Name (style):" prefix overrides this for that line.

Default: ""

voicestring

Voice for the narrator, or for the first speaker in two-speaker mode.

Default: "Kore"

AchernarAchirdAlgenibAlgiebaAlnilamAoedeAutonoeCallirrhoeCharonDespinaEnceladusErinomeFenrirGacruxIapetusKoreLaomedeiaLedaOrusPulcherrimaPuckRasalgethiSadachbiaSadaltagerSchedarSulafatUmbrielVindemiatrixZephyrZubenelgenubi

Version: ca7b5899c663Updated: 11/7/202626 runs