No audio yet. Submit the form to generate audio.
Explore different use cases and parameter configurations
Transparent pricing with no hidden fees. Pay as you go.
| Rule & Modality | Channel | Credits | Price (USD) | Official / Reference Price | Daily Savings |
|---|---|---|---|---|---|
text-to-speech otherElevenlabs | Starter | 1per prompt/1000 | $0.005 | Fal$0.100per prompt/1000 | Based on 10,000 requests/dayFal costs $954.550/day moreApiPass is 95.46% lower |
Complete guide to using ElevenLabs V3 TTS
Build expressive text-to-speech workflows with ElevenLabs V3 TTS on ApiPass. Generate natural AI voice audio from scripts, guide performance with audio tags like [excited], [whispers], and [laughs], and control voice, stability, language, text normalization, and timestamps through one async API.

ElevenLabs V3 TTS is an expressive text-to-speech model that turns written scripts into natural AI voice audio. It is designed for voice output that performs, not just reads, making it useful for narration, dialogue, product voiceovers, localization, character audio, and short-form media. The ApiPass API uses the public model name elevenlabs/elevenlabs-v3-tts, submits generation as an asynchronous task, and returns output file URLs through task polling or optional callbacks.
Generate natural speech from scripts up to 5000 characters. Inline audio tags such as [excited], [whispers], and [laughs] can guide tone and delivery for more directed voice output.
Choose from supported voice names including Rachel, Aria, Daniel, Lily, and others. Tune stability from 0 to 1 to control how consistent or variable the voice performance should be.
Use optional two-letter language_code values for multilingual workflows, apply text normalization with auto, on, or off, and request timestamp JSON output when downstream alignment data is needed.
Control emotional tone, delivery, and human reactions directly inside the script. The same text field can carry spoken content and performance cues, helping teams create narration, character dialogue, and social voiceovers from one request.
Send the documented fields text, voice, stability, timestamps, language_code, and apply_text_normalization. ApiPass validates types and supported values, then runs generation through a standard async task workflow.
Submit a task, poll /api/v1/jobs/recordInfo, or receive an optional callback. Successful tasks return resultJson.resultUrls with generated audio URLs and, when timestamps are requested and produced, additional JSON file URLs.
Generate expressive narration for reels, product demos, ads, explainers, launch videos, and social media clips using inline delivery cues.
Turn course scripts, onboarding guides, help articles, and tutorials into clear spoken audio for education platforms and internal training tools.
Prototype character lines, app assistant speech, roleplay flows, narrative dialogue, and performance-aware voice experiences using expressive tags.
Generate multilingual speech assets with optional language codes and text normalization control for global product, media, and support workflows.
ApiPass exposes ElevenLabs V3 TTS through a standard asynchronous task workflow that is consistent with other ApiPass model APIs.
Send a POST request to /api/v1/jobs/createTask with model set to elevenlabs/elevenlabs-v3-tts and an input object containing text plus optional voice, stability, timestamps, language_code, and apply_text_normalization fields.
Use the returned taskId with GET /api/v1/jobs/recordInfo to check task state. If callBackUrl is provided during creation, ApiPass will POST the final task result to your endpoint when generation completes.
When data.state is success, read data.resultJson.resultUrls. The result may be an object or a parseable JSON string, and timestamp-enabled tasks may include both audio and JSON URLs.
Current regular-channel billing is 16 credits per 1000 input characters, calculated proportionally as 16 × len(text) ÷ 1000 without rounding up.
Billing uses the raw input string length, including spaces, newlines, and punctuation. Text validation rejects empty trimmed text, but the original string is sent and counted.
voice, stability, timestamps, language_code, and apply_text_normalization do not currently change the credit formula. Timestamp files currently have no separate additional credit item.
Online billing depends on deployed and active model credit rules. Promotional pricing, channel pricing, or later migrations may change the actual credits charged.
Create an ApiPass API key from the dashboard and use it as a Bearer token when calling the task creation and query endpoints.
Prepare 1-5000 characters of text. Add expressive tags such as [excited], [whispers], or [laughs] when you want the model to follow performance direction.
Select a supported case-sensitive voice name, tune stability from 0 to 1, and optionally set language_code, timestamps, and apply_text_normalization.
Submit the task to /api/v1/jobs/createTask, then query /api/v1/jobs/recordInfo with the returned taskId. Finished tasks return audio file URLs and may include timestamp JSON URLs when requested.
All APIs require authentication via Bearer Token.
Authorization: Bearer
Submit a new ElevenLabs V3 text-to-speech task and optionally request timestamp output.
The API accepts a JSON payload with the following structure:
1{
2 "model": "elevenlabs/elevenlabs-v3-tts",
3 "callBackUrl": "string (optional)",
4 "channel": "auto",
5 "input": {
6 "text": "string",
7 "voice": "Rachel",
8 "stability": 0.5,
9 "timestamps": false,
10 "language_code": "en",
11 "apply_text_normalization": "auto"
12 }
13}modelRequiredstringThe ApiPass public model name. Must be fixed to elevenlabs/elevenlabs-v3-tts for this API.
"elevenlabs/elevenlabs-v3-tts"
callBackUrlOptionalstringCallback URL for task completion notifications. If omitted, no callback will be sent.
"https://your-domain.com/api/callback"
channelOptionalstringYou may specify the corresponding provider within APIPASS via the channel parameter. The default channel is auto. APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics. starter is ultra-low-cost with limited quotas and weaker stability. regular is the standard option and is much cheaper than official APIs with moderate stability. official uses the model native API with high stability, fast task execution, and official pricing.
The default value for the channel parameter is auto. When enabled, APIPASS automatically allocates tasks across available providers based on real-time pricing and stability metrics to balance minimal cost and reliable performance. starter is ultra-low-cost with limited quotas and weaker stability; regular is standard and much cheaper than official APIs with moderate stability; official uses the model native API with high stability, fast task execution, and official pricing.
Available options:
auto
The input object contains the following parameters:
input.textRequiredstringText to convert into speech. Must be a string between 1 and 5000 Python characters after validation, and it cannot be empty after trimming leading and trailing whitespace.
Validation uses text.strip() to reject empty text, but the original text is sent upstream without trimming. Billing uses len(text) on the raw string, so spaces, newlines, and punctuation count toward character usage. Text longer than 5000 characters is not split automatically; split long text into multiple tasks yourself.
"Welcome to ApiPass. This sentence will be converted to speech."
input.voiceOptionalstringVoice character to use for speech generation. Defaults to Rachel when omitted or set to null.
Voice names are case-sensitive exact values. Variants such as rachel or RACHEL are not converted automatically and unsupported voices do not silently fall back to Rachel.
Available options:
"Rachel"
input.stabilityOptionalnumberControls voice stability. Defaults to 0.5. Valid range is 0 to 1 inclusive.
Must be a JSON number. Integers 0 and 1 and valid floating-point values are accepted and converted to float before sending upstream. JSON booleans, numeric strings such as "0.5", NaN, and infinity values are rejected.
0.5
input.timestampsOptionalbooleanWhether to request text-to-audio timestamp information. Defaults to false.
When true, the upstream provider may return an additional timestamp JSON file alongside the audio file. The unified ApiPass resultJson extracts file URLs into resultUrls and does not retain each upstream file_type, so clients should handle multiple URLs and identify audio versus JSON by content or suffix.
Available options:
false
input.language_codeOptionalstringOptional ISO 639-1 two-letter language code, such as en, zh, ja, or fr. When omitted or set to null, this field is not sent upstream.
Available options:
"en"
input.apply_text_normalizationOptionalstringControls text normalization behavior. Defaults to auto when omitted or set to null.
Must be a string. The input is trimmed and lowercased before validation, so " ON " is normalized to "on". Other values return a parameter error and do not fall back to the default.
Available options:
"auto"
1curl -X POST "https://api.apipass.dev/api/v1/jobs/createTask" \
2 -H "Content-Type: application/json" \
3 -H "Authorization: Bearer YOUR_API_KEY" \
4 -d '{
5 "model": "elevenlabs/elevenlabs-v3-tts",
6 "callBackUrl": "https://your-domain.com/api/callback",
7 "channel": "auto",
8 "input": {
9 "text": "Hello! This is an ElevenLabs V3 text-to-speech test.",
10 "voice": "Rachel",
11 "stability": 0.5,
12 "timestamps": false,
13 "language_code": "en",
14 "apply_text_normalization": "auto"
15 }
16 }'1{
2 "code": 200,
3 "message": "success",
4 "data": {
5 "taskId": "task_xxxxxxxxxxxxxxxx"
6 }
7}codeStatus code, 200 for success, others for failure.
messageResponse message. When the request fails, this contains the error description.
data.taskIdTask ID for querying task status and results.

Starting from
0 credits
elevenlabs
elevenlabs
Create emotionally rich, directed vocal performances
Starting from
0 credits