Skip to content
HiAPI Docs
HiAPI
Playground Get API Key
English
English 中文

ElevenLabs Turbo 2.5 TTS API

POST Base URL: https://api.hiapi.ai /v1/tasks

Image, video, and audio models are called through theUnified Async API POST /v1/tasks endpoint; only the input fields differ (see input parameters below).

Model summary

Status Online
HiAPI model ID elevenlabs/text-to-speech-turbo-2-5
Price $0.043 / 1,000 characters
Text per task 1-5,000 characters
Type Audio generation (text-to-speech)

ElevenLabs Turbo 2.5 TTS is ElevenLabs' low-latency, cost-efficient text-to-speech model for voice assistants, notifications, and high-volume voiceover. HiAPI exposes 67 preset voices through the unified async task API and bills by text characters.

Production guidance

Parameter notes
  • Public input supports text, voice, stability, similarity_boost, style, speed, timestamps, previous_text, next_text, language_code.
  • Only text characters are billed ($0.043 / 1,000 characters); previous_text and next_text are free. Text over 5,000 characters is rejected at submission without charge.
  • language_code is a lowercase two-letter ISO 639-1 code; omit it for auto-detection.

Best suited for

Voice assistants and agents

Low-latency short replies and prompts.

textvoicespeed
Batch notifications

Turn reminders, order updates, and support scripts into audio at scale.

textvoice
Language-locked output

Lock the output language with language_code for mixed text.

textlanguage_code

Request parameters

model string required

Fixed value: elevenlabs/text-to-speech-turbo-2-5.

example elevenlabs/text-to-speech-turbo-2-5
input object required

Speech parameters for ElevenLabs Turbo 2.5 TTS.

text string required

Text to speak, 1-5,000 characters. Billed by text characters ($0.043 / 1,000 characters); each CJK character counts once.

voice string required

Preset voice_id; 67 voices available. View preset voices

default EkK5I93UQWFDigLMpZcX
stability number optional

0-1. Lower is more expressive, higher is more consistent.

default 0.5
similarity_boost number optional

0-1. How closely output follows the selected voice.

default 0.75
style number optional

0-1. Style exaggeration; higher adds emphasis.

default 0
speed number optional

Speaking speed, 0.7-1.2.

default 1
timestamps boolean optional

Return word-level timestamps.

default false
previous_text string optional

Text before this clip (≤5,000 characters) for continuity. Not billed.

next_text string optional

Text after this clip (≤5,000 characters) for continuity. Not billed.

language_code string optional

Optional two-letter ISO 639-1 code (e.g. zh, en, ja) to enforce the language; omit for auto-detect.

callback object optional

Optional callback configuration for terminal task notifications.

url string required

HTTPS endpoint required when callback is set.

example https://your-domain.com/hiapi/callback
when enum optional

Callback timing; use final.

default final enum: final

Preset voices (67)

Pass a voice_id from this table as input.voice.

Show all 67 voices Collapse catalog
VoiceStylevoice_id
JamesHusky, EngagingEkK5I93UQWFDigLMpZcX
ArabellaMysterious, EmotionalZ3R5wn05IrDiVCyEkUrK
BradfordExpressive, ArticulateNNl6r8mD7vthiJatiJt1
XavierBold, Metallic AnnouncerYOq2y2Up4RgXP2HyXjE5
KuonCheerful, Clear, SteadyB8gJV1IhpuegLxdpXFOE
Monika SogamDeep, Natural2zRM7PkgwBPiau2jvVXc
MarkCasual, Relaxed1SM7GgM6IMuvQlz2BwM3
AdelineFeminine, Conversational5l5f8iK3YPeGga21rQIX
SamSupport AgentscOwDtmlUjD3prqpp97I
Spuds OxleyWise, ApproachableNOpBlnGInO9m6vDvFkFC
EveGenuine, Lively, HappyBZgkqPqms7Kj9ulSkVzn
Northern Terry—wo6udizrrtpIxWGp2qJk
British Football Announcer—gU0LNdkMOQCOrPrwtbee
BrockCommanding SergeantDGzg6RaUqxGRTHSBjfgF
NathanVirtual Radio Hostx70vRnQBMBu4FAYhjJbO
AnikaLively, Friendly, EngagingSm1seazb4gs7RSlUVw7c
VirajRich, SoftP1bg08DkjqiVEzOn76yG
TakshCalm, Serious, SmoothqDuRKMlYmrm8trt5QyBn
HoratiusEnergetic CharacterqXpMhyvQqiRxWQs4qSSB
LiamEnergetic CreatorTX3LPaxmHKxFdv7VOQHJ
CallumHusky RogueN2lVS1w4EtoT3dr4eOWO
LauraWarm, QuirkyFGY2WhTYpPnrIDTdsKH5
BrittneyFun, Young, InformativekPzsL2i3teMYv0FxEYQ6
MarkNatural ConversationUgBBYS2sOqTuMpoF3BR0
BellaProfessional, Bright, Warmhpp4J3VqNfWAUOO0d1Us
BrianDeep, Resonant, ReassuringnPczCjzI2devNBz1zQrb
HopeLively, Gossipy, YouthfuluYXf8XasLslADfZ2MB4u
JeffElegant, Resonant, Stronggs0tAILXbY5DNrJrsM6F
JamahalYoung, Energetic, NaturalDTKMou8ccj1ZaWGBiotd
FinnYoung, Eager, EnergeticvBKc2FfBKJfcZNyEt1n6
TomConversation & BooksDYkrAHD8iwork3YSUBbs
CassidyCrisp, Direct, Clear56AoDkrOh6qfVPDXZ7Pt
Addison 2.0Australian AudiobookeR40ATw9ArzDf9h3v7t7
Jessica Anne BogartChatty, Friendlyg6xIsTj2HwM6VR4iXFCw
LucyFresh, CasuallcMyyd2HUfFzxdCaC4Ta
TiffanyNatural, Warm6aDn1KB0hjpdcocrUkmq
FelixWarm, Positive, Modern RPSq93GQT4X1lKDXsQcixO
Jessica Anne BogartEloquent VillainflHkNRp1BlvT73UL6gyz
LutzGiggly, Cheerful9yzdeviXkFddZ4Oz8Mok
EmmaCute, UpbeatpPdl9cQBQq4p6mRkZy2Z
EdwardLoud, Confident, CockyzYcjlYFOd3taleS0gkk3
MarshalFriendly, Witty ProfessornzeAacJi50IvxcyDnMXa
John MorganRugged CowboyruirxsoakN0GWmGNIo04
AriaSultry VillainTC0Zp7WVFzhA8zpTlRqV
Viking BjornEpic Medieval Raiderljo9gAlSqKOvF6D8sOsX
Pirate Marshal—PPzYpIqttlTYA83688JI
Johnny KidSerious, Calm Narration8JVbfL6oEdmuxKn5DK2C
HopePoetic, Romantic, CharmingiCrDUkL56s3C8sCRl7wb
Ana RitaSmooth, Expressive, BrightwJqPPQ618aTW29mptyoc
John DoeDeepEiNlNiXeDU1pqqOPrYMO
Burt Reynolds™Deep, Smooth, Clear4YYIPFl9wE5c4L2eu2Gb
HankDeep, Engaging Narration6F5Zhi321D3Oq7v1oNT4
WyattWise Country CowboyYXpFCvM1S3JbWEJhoskW
PhilExplosive, Passionate AnnouncerLG95yZDEHg6fCZdQjLqj
Johnny DynamiteRetro Radio DJCeNX9CMwmxDxUF5Q2Inm
Rachel MPro British Radio HostaD6riP1btT197c6dACmy
Rex ThunderDeep, Toughmtrellq69YZsNwzUSyXh
EdLate-Night AnnouncerdHd5gvgSOzSfduK4CvEg
JeanCharming, Playful Femme FataleeVItLK1UvXctxuaRV2Oq
BritneyCool, Calculating Villainesy0r39YPLQjOczyOib8
SvenEmotive, KindTsns2HvNFKfGiNjllgqo
VirajSmooth, Gentle1U02n4nD6AdIZ9CjF053
NathanielEngaging, British, CalmAeRdCCKzvd23BpJoofzx
BenjaminDeep, Warm, CalmLruHrtVF6PSyGItzMNHS
AllisonCalm, Soothing, Meditative1wGbFxmAM3Fgw63G1zZJ
Theodore HQSerene, ComposedhqfrgApggtO1785R4Fsn
LeonSoothing, ComposedMJ0RnG71ty4LH3dvNfSd

Example requests

Basic speech request

Default voice and settings.

Request body
{
  "model": "elevenlabs/text-to-speech-turbo-2-5",
  "input": {
    "text": "Welcome to today's product update. In the next three minutes, we will walk through what is new and how it saves you time.",
    "voice": "EkK5I93UQWFDigLMpZcX",
    "stability": 0.5,
    "similarity_boost": 0.75,
    "speed": 1,
    "language_code": "en"
  }
}
Segmented continuity request

previous_text / next_text provide context only and are not billed.

Request body
{
  "model": "elevenlabs/text-to-speech-turbo-2-5",
  "input": {
    "text": "Chapter two. The rain kept falling until dawn.",
    "voice": "EkK5I93UQWFDigLMpZcX",
    "previous_text": "At the end of chapter one, she closed the window.",
    "next_text": "The morning streets were empty.",
    "timestamps": true
  }
}

Getting the result

  1. A successful submission returns taskId immediately.
  2. Prefer callback.url in production; poll GET /v1/tasks/:id locally.
  3. After status=success, download audio from output[].url and store it as needed.

FAQ

How do I call this model?

Create an API key, send the model ID and input to POST /v1/tasks, then poll with taskId or configure a callback.

What does it cost?

$0.043 / 1,000 characters, counting text characters only.

How are CJK characters counted?

Every character, including each CJK character, counts as one.

What is the maximum text length?

5,000 characters; longer text is rejected at submission without charge.

Turbo 2.5 or Multilingual v2?

Turbo 2.5 is faster, half the price, and supports language_code; Multilingual v2 is more consistent for long-form content.

Are failed tasks charged?

Failed tasks are not billed; reserved credits are returned automatically.

Next steps