Gemini 3.8 Flash TTS API

Introduction

Gemini 3.8 Flash TTS reads a script aloud in one of 30 prebuilt voices and returns a single WAV file: 24 kHz, 16-bit, mono PCM. Send text with up to 5,000 characters, pick a voice_id (default Kore), and optionally tell it how to read with style, accent, pace, instructions and scene. For a conversation, send dialogue and speakers instead and two voices take turns in the same clip. There are no word timings or subtitles in the result.

Key features

Direction that is not spoken

style, accent and pace set the manner, English accent and rhythm; leave them out and the voice keeps its own. instructions takes free text about how to read, and scene says where and to whom the words are spoken. Neither is read aloud. temperature (0–2) sets how much the delivery varies between takes.

Acted cues inside the script

Cues in square brackets such as [whispering], [sighs], [laughing] or [gasping] are acted out rather than read aloud. Commas, ellipses and full stops set the pauses.

Two voices in one clip

speakers holds one or two entries, each with a name, a voice_id and optionally accent, style, pace and profile (a short description of who is speaking, not read aloud). Each dialogue line is {speaker, text}; lines are performed in order, up to 20 lines of 250 characters each. Settings a speaker leaves out fall back to the top-level accent, style and pace.

Length and timing

A 270-character paragraph came back as 20 to 24 seconds of audio in our tests. The model stops at 16,384 audio tokens per clip, and most tasks finish in under a minute.

Request parameters

Pricing

Prices are in credits, the same credits you use on the website (1 credit = $0.01). The hold is taken when the task is accepted and settled to the actual charge when it finishes; failed tasks are refunded.

For an exact number before you run, send the same body to POST /jobs/quote.

Use cases

Narration for videos and explainers

Write the voiceover, choose a voice and a pace, and put the tone in instructions ("warm, unhurried, like a museum audio guide"). The WAV drops straight into an editor.

Two-person dialogue

Podcast intros, product demos and scripted interviews fit the two-speaker format: give each speaker a profile and a voice, and write the exchange as dialogue lines.

Characters with a performance

Bracketed cues let a line be whispered, laughed through or sighed, which makes game barks, audiobook characters and short sketches sound acted rather than read.

Prompt guide & pro tips

Pick the voice by character, not by language

Each voice has a fixed character, for example Kore firm, Charon informative, Puck upbeat, Achernar soft and Sulafat warm, and it keeps that character in any language the script is written in.

Say the scene, then the manner

scene answers "where and to whom" ("a late-night radio show, talking to one listener"); instructions answers "how" ("slow, with a smile in the voice"). Keeping the two apart gives steadier takes than one long description.

Let punctuation do the timing

An ellipsis adds a breath, a full stop ends a thought, a short sentence speeds things up. Fix pacing in the script before reaching for pace.

Keep long scripts in paragraphs

Split long reads into separate tasks at natural paragraph breaks; it keeps each take short enough to redo without regenerating the whole piece.

API quick start

Call the Gemini 3.8 Flash TTS API in three steps

1. Create an API key. Sign in to PopNow and create a key in the API console. Keys start with sk-popnow- and use the same credits as the website.
2. Send the request. POST /jobs/createTask with model google/gemini-3.8-flash-tts/text-to-speech and an input object that follows the parameter table. The response returns a task ID straight away.
3. Get the result. Poll GET /jobs/recordInfo with the task ID, or pass a callBackUrl and we post the result to it. Download the output links when the task succeeds.

# 1. Create the task. Keep the Idempotency-Key if you retry after a network error.
export IDEMPOTENCY_KEY="$(uuidgen)"
curl -X POST https://api.popnow.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  -d '{
  "model": "google/gemini-3.8-flash-tts/text-to-speech",
  "input": {
    "text": "Welcome back. Your files finished rendering a minute ago and are ready to download.",
    "voice_id": "Kore"
  }
}'

# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $POPNOW_API_KEY"

Replace the prompt and any links with your own. The API tab has the response, polling and callback reference.

Gemini 3.8 Flash TTS API questions

All Gemini models

Gemini 3.8 Flash Lite TTS uses the same voices for less, and suits drafts and long batches. The full Flash model gives the more expressive read.

• Gemini 3.8 Flash TTS APIExpressive speech that acts out cues like [laughing]From 0.05 credits per second
• Gemini 3.8 Flash-Lite TTS APIFast, lightweight speech in 30 voicesFrom 0.03 credits per second
• Gemini 2.5 Pro TTS APIDirectable voices for narration and two-person dialogueFrom 0.07 credits per second

Other audio APIs

• MiniMax Speech 2.8 HD APIHigh-fidelity speech in 400+ voices with emotion controlFrom 18 credits per 1,000 characters
• Grok TTS API71 voices across 20 languages with speech tagsFrom 3 credits per 1,000 characters
• Seed Speech 2.0 API41 voices with tone set in plain languageFrom 11 credits per 1,000 characters
• MiniMax Speech 2.8 Turbo APIFast speech in 400+ voices with emotion controlFrom 11 credits per 1,000 characters
• Qwen3 TTS APIOpen-source speech with tone and pace set in plain wordsFrom 11 credits per 1,000 characters
• Chatterbox APIEnglish speech that can match a short voice sampleFrom 6 credits per run

More about Gemini 3.8 Flash TTS

Try it

Gemini 3.8 Flash TTS AI Voice Generator onlineCreate audio with Gemini 3.8 Flash TTS in PopNow

Docs

Gemini 3.8 Flash TTS model ID and input schema in the API docsPopNow API quickstart: your first request in five minutesWebhooks: get Gemini 3.8 Flash TTS results without polling