Gemini 3.8 Flash Tts

A Google speech model for expressive narration, audiobooks and dialogues with voice control and support for 130 languages.

Ailoot · Text to speech POST /v1/audio/speech gemini-3.8-flash-tts

About the model

This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.

Gemini 3.8 Flash TTS is intended for expressive narration and long-form storytelling. Google highlights audiobooks, multi-speaker dialogues and voice consistency over extended passages. Speech control covers emotional delivery, rhythm and pronunciation, including regional accents and dialects.

Voices and languages

The model is described as supporting 130 languages, including Russian. 30 preset voices, an expanded library, creation of a voice identity from a description and reproduction of a voice from a sample are available. These capabilities allow timbre and performance character to be selected for particular material.

Dialogue lines are associated with speakers. One multi-speaker request can use no more than two preset voices. For voices created from a description or sample, lines are synthesized separately; having such a voice does not make it another preset voice within a multi-voice request.

Text and delivery control

Input is text, and output audio. Limits are 8192 input tokens and 16,384 output tokens. Text is treated as a spoken line; style description and the selected speaker are specified separately. This separates the words to be spoken from instructions about how to read them.

Style covers emotion, pace, volume and prosody. Brief events such as laughter, sighs and pauses are specified with special audio tags within the line. These tags control the performance rather than serving as ordinary words to be read aloud.

Audio formats and capabilities

Ordinary generation returns WAV by default. Streaming uses headerless PCM by default. Original parameters are 24 kHz, one channel and 16 bits. PCM, mu-law and A-law are also provided. Context caching and batch processing are supported. The specific file format and available settings depend on the selected speech method.

The model specializes in speech creation. Function calling, code execution, file search, search grounding, Google Maps, image generation, structured text responses, Thinking and URL contents are not supported. Consistent voice and controllable expressiveness are the model's stated aims, not a promise of perfect pronunciation in every line.

Ailoot · Text to speech POST /v1/audio/speech gemini-3.8-flash-tts

gemini-3.8-flash-tts

Google /gemini-3.8-flash-tts

Overview

A Google speech model for expressive narration, audiobooks and dialogues with voice control and support for 130 languages.

Accepted inputs: Text
Results: Audio

Available in Ailoot · inputs & methods

Ailoot · Text to speech POST /v1/audio/speech

Input types: text
Output types: audio
Context: 8,192 tokens
Maximum output: 16,384 tokens
Output format: native audio bytes or JSON asset links

Field	Type	Requirement	Values & limits	Default
Field`model`	Type`string`	RequirementRequired	Values & limits gemini-3.8-flash-tts	Default—
Field`input`	Type`string`	RequirementRequired	Values & limits Minimum length: 1	Default—
Field`voice`	Type`string`	RequirementOptional	Values & limits Achernar, Achird, Algenib, Algieba, Alnilam, Aoede, Arlo, Autonoe, Bodi, Brio, Callirrhoe, Charon, Cleo, Cruz, Daro, Despina, Elio, Enceladus, Enya, Enzo, Erinome, Fenrir, Finn, Fola, Gacrux, Gero, Hali, Iapetus, Jett, Jori, Kira, Knox, Koda, Kore, Laomedeia, Leda, Lora, Ludo, Lumi, Mako, Milo, Neno, Nika, Nyla, Olin, Orus, Puck, Pulcherrima, Rami, Rasalgethi, Riko, Rina, Sadachbia, Sadaltager, Sami, Schedar, Sola, Sulafat, Tari, Tavi, Tova, Umbriel, Varo, Veda, Vindemiatrix, Zali, Zeno, Zephyr, Zubenelgenubi, Zuri	DefaultKore
Field`response_format`	Type`string`	RequirementOptional	Values & limits , wav, json	Default
Field`stream_format`	Type`string`	RequirementOptional	Values & limits audio	Default—

Input material combinations

text_to_speech · Required: input

Method parameters and lifecycle

Parameter	Contract
Parameter`material_types`	Contracttext
Parameter`required`	Contractnull
Parameter`optional_controls`	Contractnull

Lifecycle: synchronous_result_then_owned_library

GET /v1/media/tasks/{task_id} · actual returned task_id or media library record

GET /v1/media/assets/{asset_id}/content

GET /v1/media/assets

GET /v1/media/assets/{asset_id}

Method shapes and IDs are exact; supplier capability is separate from admission.

Pricing

Ailoot method rates

Ailoot Text to speechPOST /v1/audio/speech

Input: $0.0005 / 1K
Output: $0.009 / 1K
Cache read: $0.000125 / 1K

Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →

Example requests

Ailoot · Text to speech

Speech example

set -eu
: "${AILOOT_API_KEY:?Set your Ailoot API key}"
: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 narration script}"
export AILOOT_TEXT_FILE
python3 - <<'PY_REQUEST'
import json,os
from pathlib import Path
p=json.loads('{"model": "gemini-3.8-flash-tts", "input": ""}')
f=Path(os.environ["AILOOT_TEXT_FILE"])
if not f.is_file():raise SystemExit("Choose your actual UTF-8 source material")
text=f.read_text(encoding="utf-8")
if not text.strip():raise SystemExit("Source material must contain text")
p["input"]=text
Path("request.json").write_text(json.dumps(p,ensure_ascii=False))
PY_REQUEST
curl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/speech \
  -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \
  --data-binary @request.json -D response-headers.txt -o speech.wav

Receive and save results

Ailoot · Text to speech

native audio bytes or JSON asset links

GET /v1/media/tasks/{task_id}
GET /v1/media/assets/{asset_id}/content
GET /v1/media/assets
GET /v1/media/assets/{asset_id}

Keep all completed results

Keep every complete returned response, file and content part.

A single output defaults to native audio bytes with actualContent-Type; response_format=json or multiple outputs returns all owned links. No requested MP3/Opus/AAC/FLAC/PCM conversion or SSE.

If the original response needs checking

Retain original operation_id/X-Client-Request-ID for reconciliation without resubmission; no dedicated audio poll route.

Response fields

Field	Type
Field`binary native audio`	Typebytes
Field`model`	Typestring
Field`response_id`	Typestring
Field`actual_count`	Typeinteger
Field`data[]`	Typeowned audio asset link

Errors and next steps

Response	Condition	Next step
Response400 · `invalid_request_error`	ConditionInvalid required fields, material representation or parameter value	Next stepCorrect the indicated fields before a new creation request.
Response`server_error`	ConditionThe returned response says the operation or result requires a check	Next stepKeep the original returned operation/task ID, request headers and response; continue supported reads of that original operation. Do not resubmit the creation request.

Aliases

Model aliases

Ailoot gemini-3.8-flash-tts

Method path aliases

Use the method paths shown above.

Related caller IDs

gemini-3.8-flash-tts

FAQ · Gemini 3.8 Flash Tts

What inputs and results are available?: POST /v1/audio/speech · gemini-3.8-flash-tts: text → audio
Which fields does the request use?: POST /v1/audio/speech · gemini-3.8-flash-tts: model, input, voice, response_format, stream_format
What limits apply?: POST /v1/audio/speech · gemini-3.8-flash-tts: ctx: 8192; max_out: 16384

Gemini 3.8 Flash Tts