Supertonic 3 Text to Speech — 31 Languages, Studio-Grade AI Voices ($10 per 1M characters)

Turn any text into natural, human-like speech in 31 languages — one flat all-inclusive price, no subscriptions, no hidden platform fees.

Built for AI agents, LLM pipelines, n8n / Zapier / Make automations, app prototyping and content production. Pay only for successfully delivered audio.

Table of contents

• What can it do
• Who uses it
• Input fields
• How to use
• Output example
• Pricing
• Integrations and API
• Tips
• FAQ
• Related searches
• Disclaimer
• 中文说明

What can it do

✅ Convert plain text (1–2,000 Unicode code points per run) into natural speech ✅ 31 supported languages including English, Turkish, German, French, Spanish, Japanese, Korean, Arabic, Russian, Portuguese, Italian, Dutch, Polish, Ukrainian and more ✅ 10 official preset voices — five female (F1–F5) and five male (M1–M5) ✅ Adjustable speaking speed (0.7×–1.5×) ✅ Returns a WAV audio clip plus SHA-256 checksum, byte size and expiry timestamp ✅ One flat all-inclusive price per started 1,000 characters — no start fee, no platform surcharge ✅ Charges only on successful, non-empty delivery — failed runs cost nothing ✅ Works with AI agents and LLM tools via standardized input/output

Who uses it

• AI agent developers giving assistants a spoken interface
• Automation builders on n8n, Zapier, Make and Windmill generating voice notifications, IVR prompts or podcast drafts
• App developers prototyping voice features without committing to a big-cloud subscription
• Content creators producing narration drafts in multiple languages
• Accessibility teams adding on-demand speech to reading flows

Input fields

Field	Type	Required	Description
text	string	✅	Plain text to synthesize, 1–2,000 Unicode code points. Billing counts code points including spaces and punctuation, in started blocks of 1,000 ($0.01 per block, minimum one block).
voice	select	❌	Official preset voice ID: F1–F5 (female) or M1–M5 (male). Defaults to F1. No voice cloning.
language	select	❌	One of the 31 supported Supertonic-3 languages. Defaults to en.
speed	number	❌	Speaking-rate multiplier, 0.7–1.5. Defaults to 1.05.
acceptTerms	boolean	✅	Must be true: you agree to disclose synthetic speech and not impersonate or mislead.

How to use

1. Add your text — paste or send 1–2,000 characters.
2. Pick a voice — try F1 for a warm female narrator or M2 for a calm male voice.
3. Choose language and speed — e.g. tr for Turkish at 1.0.
4. Accept the synthetic-speech terms — set acceptTerms: true.
5. Run — the actor returns a signed HTTPS download URL (valid 24 hours), audio metadata and billing details in the Output tab and dataset.

Input example

{
  "text": "Welcome to our service. This voice was generated by AI.",
  "voice": "F1",
  "language": "en",
  "speed": 1.05,
  "acceptTerms": true
}

Output example

Real run output (abridged):

{
  "ok": true,
  "kind": "tts",
  "voice": "F1",
  "language": "en",
  "charCount": 49,
  "duration": 4.2,
  "audioUrl": "https://minio.foxtools.de/...?X-Amz-Signature=...",
  "audioDelivery": "minio-presigned",
  "deliveryObjectKey": "foxo81/audio/....wav",
  "audioBytes": 370720,
  "audioSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
  "expiresAt": "2026-09-19T12:00:00Z",
  "billingUnit": "text-1k-chars",
  "billingQuantity": 49,
  "billingCount": 1,
  "billingPriceUsdPerBlock": 0.01,
  "totalChargeUsd": 0.01,
  "billingIsAllInclusive": true
}

Pricing

$0.01 per started 1,000 characters — all-inclusive.

The math, shown plainly:

Characters	Blocks (ceil(chars/1000))	Cost
1–1,000	1	$0.01
1,001–2,000	2	$0.02
500,000	500	$5.00
1,000,000	1,000	$10.00

That equals $10 per 1M characters — below most major cloud TTS APIs. One price covers speech synthesis, Apify platform usage, storage and audio delivery. No separate start fee, no usage surcharge. Blocks are charged only for a successfully delivered, non-empty clip; failed runs cost $0.

Integrations and API

Sync run (get dataset items directly):

curl -X POST "https://api.apify.com/v2/acts/myagizm~supertonic-3-text-to-speech/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"text":"Hello from the API!","voice":"F1","language":"en","speed":1.05,"acceptTerms":true}'

Async run:

curl -X POST "https://api.apify.com/v2/acts/myagizm~supertonic-3-text-to-speech/runs?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"text":"Async speech generation.","acceptTerms":true}'

n8n / Zapier / Make: use the Apify node, pick this actor, map your text field to text and set acceptTerms: true.

Tips

💡 Batch your scripts: 2,000 characters per run is the sweet spot — exactly 2 blocks. 💡 Download the WAV before expiresAt (24 h) if you need a permanent copy. 💡 Verify integrity with audioSha256 after downloading. 💡 Use speed: 0.9 for tutorials, 1.2 for energetic promos.

FAQ

Q: Which languages are supported? English, Turkish, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Ukrainian, Russian, Arabic, Japanese, Korean, Bulgarian, Czech, Danish, Greek, Estonian, Finnish, Hindi, Croatian, Hungarian, Indonesian, Lithuanian, Latvian, Romanian, Slovak, Slovenian, Swedish, Vietnamese.

Q: Can I clone a voice? No. Only the ten official presets F1–F5 / M1–M5 are offered — this keeps output quality consistent.

Q: What audio format do I get? Uncompressed PCM WAV via a signed HTTPS download URL, valid for 24 hours.

Q: Do failed runs cost money? No. A block is charged only for a successfully delivered, non-empty clip.

Q: Is the audio watermarked? No watermark, but output is AI-generated and must be disclosed as such.

Q: Why Unicode code points? One emoji or non-Latin glyph counts as one code point, matching how models actually read text.

Q: Can I use this commercially? Yes — audiobooks, apps, ads — as long as you disclose synthetic speech and do not impersonate real people.

Q: How fast is a run? Typically a few seconds for short texts; the actor's internal deadline is 420 s.

Q: Is there a free trial? Apify's platform free tier gives you platform credits; per-event charges apply per delivered clip.

Q: What about long documents? Split them into ≤2,000-character chunks and run per chunk.

Related searches

supertonic text to speech api · ai voice generator api · cheap tts api · multilingual tts api · turkish text to speech api · german tts api · ai speech synthesis · natural sounding tts · tts for n8n · tts for zapier · ai voiceover api · flat price tts · pay per use tts · tts actor apify · speech synthesis 31 languages

Disclaimer

Output is AI-generated speech. You are responsible for disclosing synthetic audio and for not using it to impersonate, defraud or mislead. We do not claim quality equivalence with any third-party service.

中文说明

Supertonic 3 文本转语音 — 31 种语言，$10 / 100 万字符

把任意文本变成自然的 AI 语音 — 一个全包价，无订阅、无隐藏平台费。

✅ 31 种语言（英语、土耳其语、德语、法语、西班牙语、日语、韩语、阿拉伯语等） ✅ 10 个官方预设音色：F1–F5（女声）、M1–M5（男声） ✅ 语速 0.7–1.5 可调 ✅ 返回 WAV 音频 + SHA-256 校验 + 24 小时下载链接 ✅ 计费：每 1,000 字符 $0.01（起步一块），失败不收费

输入字段

字段	类型	必填	说明
text	string	✅	纯文本，1–2000 Unicode 字符
voice	select	❌	F1–F5 / M1–M5，默认 F1
language	select	❌	31 种语言之一，默认 en
speed	number	❌	0.7–1.5，默认 1.05
acceptTerms	boolean	✅	必须为 true：同意披露合成语音

计费示例

字符数	块数	费用
1–1,000	1	$0.01
1,001–2,000	2	$0.02
1,000,000	1,000	$10.00

相关搜索

文本转语音 API · AI 配音 API · 多语言 TTS · 土耳其语 TTS · n8n 语音 · 便宜 TTS API