gemini-3.8-flash-lite-tts

Text to Speech

Gemini 3.8 Flash Lite TTS brings multilingual speech to growing content libraries and voice-enabled apps, with support for 101 languages, 30 preset voices, and two-speaker dialogue. On PoYo, keep spoken text separate from style instructions to create narrated articles, online lessons, and audio content with a suitable tone and clearly assigned speakers.

Get APIKey

API Doc

Input

Text

Voice

Speakers

Turns

Style Instructions

Enable Safety Checker

Billing includes text and all style instructions.

326 characters

Output

Example Output

This is sample data. Generate a new result to see real output.

Task ID:example-gemini-3-8-flash-lite-tts-defaultCreated:2026-10-14T11:13:12

Status:finishedProgress:100%

Examples

Pricing details

Transparent pricing with no hidden fees. Pay as you go.

gemini-3.8-flash-lite-tts

Text and style instructions

PoYo price

$0.030

6 credits

Official price

You save

Model	Spec	PoYo price	Official price	You save
gemini-3.8-flash-lite-tts	Text and style instructions	$0.030 6 credits per 1000 characters	-	-

* Actual fees are based on the final output.

GEMINI 3.8 FLASH LITE TTS

Multilingual narration and dialogue for voice applications

Gemini 3.8 Flash Lite TTS API for Natural Speech Synthesis at Scale

Make articles, lessons, and a growing content library ready to listen to. With Gemini 3.8 Flash Lite TTS on PoYo, choose preset voices and describe the delivery style to create multilingual narration and two-speaker dialogue for reading apps and content platforms.

Get API Key

Key Features of Gemini 3.8 Flash Lite TTS API

Guide speech delivery, stage two-speaker turns, and create multilingual narration.

Natural Speech with Style Instructions

A useful voiceover needs the right delivery as well as the right words. Keep your transcript in text and describe tone, pacing and expression separately in style_instructions. Gemini 3.8 Flash Lite TTS lets you adapt a read for a friendly introduction, a calm explanation or a more expressive scene, while keeping performance direction separate from the spoken script.

Describe tone and pacing in separate instructions
Choose from 30 supported preset voices
Keep spoken words separate from delivery direction

Structured Dialogue with Two Distinct Voices

Give podcast segments or teaching Q&A a clear speaker structure. Configure two speakers in speakers, then let each turn in turns reference the corresponding speaker_id. Optional per-turn style_instructions can adjust each line's delivery, while revisions keep speaker roles and dialogue order clear.

Assign two speaker IDs and preset voices
Write dialogue as ordered structured turns
Direct each turn with optional style instructions

Multilingual Speech with Automatic Language Detection

Reach listeners in their own language without maintaining a separate language parameter for each request. Gemini 3.8 Flash Lite TTS supports 101 languages and detects the language from your text. Select a supported preset voice and prepare localized scripts for lessons, product walkthroughs or read-aloud experiences. The official Japanese audio illustrates multilingual speech within the Gemini 3.8 series.

Generate speech across 101 supported languages
Detect the language from the submitted text
Prepare localized narration for global audiences

Who Can Benefit from Gemini 3.8 Flash Lite TTS API?

Narration and Audiobooks

Publishers and audio content teams can turn passages into narration with Gemini 3.8 Flash Lite TTS. Choose a preset voice, keep delivery instructions separate from the story, and revise individual passages as the script changes. A structured text-to-speech workflow helps manage recurring narration and larger collections of spoken content.

Video Voiceovers

Video editors can prepare spoken tracks for explainers, product walkthroughs and social content. Use Gemini 3.8 Flash Lite TTS to generate a voiceover from the approved script, then download the completed audio into the editing workflow. Separate style instructions help adapt the delivery to each scene or audience.

Education and Training

Learning platforms and training teams can create spoken explanations, reading exercises and two-person demonstrations with Gemini 3.8 Flash Lite TTS. Set a clear delivery style for narration or use structured turns for a teacher-and-learner exchange. Localized scripts help the same lesson reach learners in different languages.

Podcast Dialogue

Podcast producers can stage scripted host-and-guest segments with two preset voices. Configure both speakers in Gemini 3.8 Flash Lite TTS, write their turns in sequence and adjust direction for individual lines. The explicit dialogue structure makes it easier to revise a question or answer without rewriting an entire transcript.

Read-Aloud Experiences

Reading applications and accessibility teams can offer spoken versions of articles, instructions and everyday content. Gemini 3.8 Flash Lite TTS supports multilingual read-aloud workflows with selectable voices and separate pacing instructions. Applications can submit prepared text, track the task and make the completed audio available to the listener.

Speech for Voice Applications

Developers building voice applications can convert prepared text responses into speech with Gemini 3.8 Flash Lite TTS. Use a consistent preset voice and delivery style for product updates, guidance and conversational replies. Integrate the asynchronous task into the application's own response pipeline and retrieve audio when generation finishes.

Gemini 3.8 Flash Lite TTS vs Gemini 3.8 Flash TTS — Model Comparison

Choose the Gemini 3.8 TTS model that fits your workload. Flash Lite emphasizes efficient speech production at scale; Flash emphasizes expressive creative performance. Both expose the same core speech controls through PoYo.

Feature	Gemini 3.8 Flash Lite TTS	Gemini 3.8 Flash TTS
Primary focus	High-throughput, cost-efficient speech production	Expressive creative production and acting nuance
Supported languages	101 languages	130 languages
Preset voices on PoYo	30 supported voices	30 supported voices
Two-speaker dialogue	Exactly two speakers with structured turns	Exactly two speakers with structured turns
Style control on PoYo	Global and per-turn style_instructions	Global and per-turn style_instructions

Positioning and language coverage follow Google's model documentation. Input controls reflect PoYo's current public backend. Official series demos are illustrative and do not measure PoYo latency.

How to Use Gemini 3.8 Flash Lite TTS API on PoYo

Step 1: Register and Create an API Key
Sign in to PoYo and create a key for your application.
Create API Key →

Step 2: Add Credits
Fund your account, then check the pricing table and Playground estimate for your script and style instructions.
Add Credits →

Step 3: Submit and Retrieve Audio
Send model: gemini-3.8-flash-lite-tts and a single-speaker or structured dialogue input. Poll the returned task ID or supply a callback URL, then download the audio when the task finishes.
View API Reference →

Frequently Asked Questions about Gemini 3.8 Flash Lite TTS API

What is Gemini 3.8 Flash Lite TTS API?

Gemini 3.8 Flash Lite TTS is Google's text-to-speech model for efficient, high-volume speech production. It generates natural multilingual audio with voice and style controls. PoYo provides asynchronous API access with 30 preset voices, single-speaker narration and structured two-speaker dialogue, plus a Playground for preparing requests and retrieving completed audio.

How is Gemini 3.8 Flash Lite TTS billed on PoYo?

The rate is 6 credits, or $0.03, per 1,000 billable characters. Billing includes single-speaker text or all turn texts, plus global and per-turn style instructions. Speaker IDs and voice names are excluded. Spaces, newlines and inline vocal tags count. Charges are proportional to character count rather than rounded up to full blocks of 1,000.

Which voices can I select?

PoYo supports 30 Gemini preset voices, including Kore, Charon, Puck, Aoede and Zephyr. Single-speaker requests default to Kore when voice is omitted. For dialogue, assign a supported voice to each of the two speakers. Keep voice names unchanged; custom voice IDs and voice-clone creation are not exposed in this integration.

Which languages are supported?

Gemini 3.8 Flash Lite TTS supports 101 languages, including English, Chinese, Japanese, Korean, Spanish and French. Language is detected from the submitted text, so PoYo does not require or accept a language_code field for this model. Prepare the script in the desired language and review pronunciation of names or specialized terms in the completed audio.

How do I control tone, pacing and expression?

Put sustained delivery instructions in style_instructions, separate from the transcript. For example, ask for a warm, clear explanation or a measured reading pace. In dialogue, you can supply global direction and optional instructions for each turn. Inline angle-bracket tags such as <short pause> can describe momentary events. All text and style instructions contribute to billing.

How do two-speaker requests work?

Configure speakers with exactly two unique, non-blank speaker_id values and supported voices. Add one or more turns, each containing a matching speaker_id and non-blank text; per-turn style_instructions are optional. Omit top-level text and voice in dialogue mode. Speaker IDs are matched exactly and may contain spaces; they are not included in the billable character count.

Can I choose an audio output format?

The current PoYo backend does not expose an output_format parameter for this model. Retrieve the audio file supplied by the completed task and preserve its original format when saving it. The Playground provides browser playback and downloads for returned audio. Its download filename follows the response MIME type or file URL rather than assuming every result is MP3.

Does PoYo provide streaming for this model?

This integration uses asynchronous task submission and status queries. Submit once, save data.task_id and query the existing task until its status is finished or failed; you may also supply callback_url outside input. Temporary status-query errors should be retried without creating another task. Streaming endpoints and realtime audio chunks are not exposed by this PoYo integration.

Why Use PoYo for Gemini 3.8 Flash Lite TTS API Access

Unified API Access

Use PoYo's existing submit and status endpoints for Gemini 3.8 Flash Lite TTS. Keep authentication and task handling within the same integration used for other supported models.

Usage-Based Billing

Review the character-based estimate before submitting. The Playground includes transcript and style instructions in the total so you can understand how each request is billed.

A Playground for Request Preparation

Configure narration or dialogue through the form, or edit the complete request as JSON. Validate the same supported fields before connecting the model to your application.

Asynchronous Task Management

Track generation with a task ID, retry temporary status-query failures and collect finished audio. Supply a callback URL when your application needs a result notification.