Thai audiobooks and voiced novels
Narrate audiobooks and voiced novels in Thai from the manuscript, one request per chapter, billed by the character with one voice across the series.
Ideas to try
Every card plays a real clip generated from its exact script with Nom Yen. Open one in the playground to make it say anything.
How a title becomes audio
Thailand already has somewhere to publish an audiobook. MEB runs a dedicated audiobook storefront, Sonar publishes Thai audiobooks, and TK Park lends them to library members at no charge. The constraint sits earlier, in the recording. Eight hours of finished audio is eight hours of studio time and the editing around it. A narrator has to be booked and held for a whole series, and a corrected sentence in chapter nine brings that voice back. A text to speech API prices the same book by its characters instead. A chapter renders while the next one is still with the editor, the narrator is a voice id that stays available for the whole series, and a corrected line costs one request.
- Split the manuscript at chapter breaksA chapter longer than 5,000 characters is sent in parts, split at paragraph boundaries. Keeping the split on the manuscript's own structure means a reissued chapter maps to the same parts.
- Render the book as a batchEvery part is one request carrying the same voice id, and the returned MP3 files join into a chapter track. A book runs unattended, and the plan's rate limit sets how fast.
- Correct a line without a sessionAn edited paragraph re-renders as one request in the same voice. The rest of the book is untouched, and the charge is the characters that changed.
A book, priced by the character
| Length | Characters | Credits | Cost |
|---|---|---|---|
| A 20-minute chapter | 12,150 | 182.25 | $0.18 |
| A series of 5 books | 1,458,000 | 21,870 | $21.87 |
| An 8-hour audiobook, narrated end to end | about $4.37 | ||
Computed at 608 characters per minute of speech at the flat rate.
For developers
The Python and TypeScript quickstarts post a chapter part and write the MP3 a batch job collects. The options request shows the voice, model, format, and stream fields a book pins once and reuses to the last chapter.
import os
import requests
response = requests.post(
"https://api.paxalabs.com/v1/tts",
headers={"Authorization": f"Bearer {os.environ['PAXA_API_KEY']}"},
json={"text": "สวัสดีครับ ยินดีต้อนรับสู่ Paxa Labs", "voice": "khanomkrok", "model": "paxa-tts-flash-v1"},
)
response.raise_for_status()
with open("speech.mp3", "wb") as f:
f.write(response.content)import { writeFile } from "node:fs/promises";
const response = await fetch("https://api.paxalabs.com/v1/tts", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.PAXA_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
text: "สวัสดีครับ ยินดีต้อนรับสู่ Paxa Labs",
voice: "khanomkrok",
model: "paxa-tts-flash-v1",
}),
});
if (!response.ok) throw new Error(`TTS failed: ${response.status}`);
await writeFile("speech.mp3", Buffer.from(await response.arrayBuffer()));curl -X POST https://api.paxalabs.com/v1/tts \
-H "Authorization: Bearer $PAXA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"text": "สวัสดีครับ ยินดีต้อนรับสู่ Paxa Labs",
"voice": "khanomkrok",
"model": "paxa-tts-flash-v1",
"format": "mp3",
"stream": false
}' \
--output speech.mp3Hear the voices
PanangMale0:03
Polished, theatrical delivery with prestige-drama gravitas.
ค่ำคืนนี้ เรื่องราวทั้งหมดจะค่อย ๆ เผยออกมา ทีละฉาก ทีละบรรทัด
Khanom KrokMale0:03
Calm, unhurried elder voice and the roster's male lead: documentary, heritage storytelling, and narration.
สมัยผมยังหนุ่ม แถวนี้ยังเป็นทุ่งนา เดินไปตลาดทีหนึ่งก็ค่อนวันแล้วครับ
Nom YenFemale0:06
Bright, energetic voice and the roster's female lead: promos, social clips, and everyday product speech.
ยินดีต้อนรับสู่ Paxa Labs ค่ะ เสียงภาษาไทยที่สลับเป็น English ได้กลางประโยค อย่างเป็นธรรมชาติ
TakoMale0:05
Friendly young narrator for audiobooks, recaps, and explainers.
ยามเช้าในกรุงเทพฯ แสงแดดอ่อน ๆ ทาบลงบนถนนที่เริ่มคึกคัก
A few of the 26 production voices, generated from the API with the published voice ids. The full roster plays on the Text-to-Speech page.
Built for book-length Thai
A chapter is thousands of characters with no spaces between the words, and a novel carries dialogue, names, and numbers on every page. Paxa Flash segments long Thai prose, reads tone from spelling the way a literate reader does, and holds one delivery across a whole series.
Common questions
- How long can one chapter be?
- Up to 5,000 characters per request. A longer chapter is sent in parts split at paragraph boundaries, and standard tools join the MP3 parts into one track.
- Will the narrator sound the same in chapter forty?
- Yes. A voice is a fixed id in the catalog. The same id returns the same voice on the last chapter as on the first.
- Can the book read Thai and English in one sentence?
- Yes. One voice carries both. An English title or name inside a Thai sentence stays in the same narrator's delivery.
- What does a full audiobook cost?
- About $4.37 for 8 hours of finished audio, billed by the characters in the manuscript. Re-rendering a corrected chapter is charged the same way.
- Do we hold the rights to publish the audio?
- Output made on free credits is for personal use and evaluation. Once you buy a plan or credits, generated output becomes yours for any use, including commercial publication, and that covers output generated before the purchase. The rights in the underlying manuscript remain yours to clear.
Start with free credits
Sign in with Google, GitHub, or Hugging Face and spend 100 one-time free credits on your own text.