Paxa Labs and ElevenLabs for English, Thai, and Mandarin speech

How a speech API for English, Thai, and Mandarin compares with a platform covering more than 90 languages, across voices, code-switching, billing, and the cases each one suits.

ElevenLabs is a multilingual speech platform whose newest models list more than 90 languages, with voice cloning at the centre of it. Paxa TTS Flash v2 speaks English, Thai, and Mandarin from a roster of 50 voices, and its design constraint is switching between Thai and English inside a single sentence. If your product ships in a dozen locales, their breadth is the point. If it speaks English, Thai, or Mandarin, the voices themselves and the rate at your volume are what to compare.

Sources:ElevenLabs models(opens in a new tab)

Rates and claims last checked October 2026

Start free See full pricing

What differs

What differs
Dimension	Paxa Labs	ElevenLabs(opens in a new tab)
What the model is built around	Paxa TTS Flash v2 speaks English, Thai, and Mandarin, and is built for Thai sentences that carry English words.	A multilingual platform whose newest speech models list more than 90 languages.
Thai and English in one sentence	Handled inside a single utterance, which is how Thai product names, brand names, and technical terms are actually spoken.	Their documentation is the place to check how they handle mid-sentence language changes.
Thai voices	All 50 voices speak Thai, each with a published id and a fixed sample line. Paxa TTS Flash, model paxa-tts-flash-v1, carries 4 more in 3 regional registers.	Voices come from their library, from cloning, and from text-prompted design; the count that matters to you is the Thai one, and their site is where to count it.
English speech	All 50 voices speak English, billed at the same per-character rate as Thai and Mandarin.	English, with US, UK, Australian, and Canadian variants named in their model documentation.
Mandarin speech	All 50 voices speak Mandarin from Simplified Chinese text, in the same voice they use for English and Thai.	Mandarin Chinese, among the more than 90 languages listed for their newest speech models.
What a failed request costs	Nothing. Credits are charged before inference and refunded automatically when generation fails, including mid-stream.	Their billing terms are on their pricing page.
Moving an existing integration	An OpenAI-compatible endpoint takes an existing SDK at a different base URL. The client code stays as it is.	They publish their own SDKs and API, which an integration is written against.

Sources:ElevenLabs models(opens in a new tab)ElevenLabs voices(opens in a new tab)ElevenLabs API pricing(opens in a new tab)ElevenLabs API reference(opens in a new tab)

Paxa TTS Flash v2, per 1M characters: $10
ElevenLabs: $40-80
Their rate, as a multiple: 4-8×

Both figures are published list rates per million characters, read from each vendor's own pricing page and re-checked on a fixed schedule. Every rate is per one million characters of English text, counted in UTF-16 code units as ours is, with per-minute, per-token, per-byte, and per-Han-character rates converted to that unit.

What it costs at your volume

What it costs at your volume
Speech, per month	Paxa Labs	ElevenLabs	Difference
100,000About 111 minutes	Paxa Labs$1	ElevenLabs$4-8	Difference$3 less
1,000,000About 18 hours	Paxa Labs$10	ElevenLabs$40-80	Difference$30 less
5,000,000About 92 hours	Paxa Labs$50	ElevenLabs$200-400	Difference$150 less
20,000,000About 371 hours	Paxa Labs$200	ElevenLabs$800-1,600	Difference$600 less

Every figure is the two published rates multiplied by the volume beside it. Where the other vendor publishes a range, the low end is compared and the range is shown.

When ElevenLabs is the better choice

If you need many languages from one vendor, ElevenLabs lists more than 90 and we serve three. A product shipping in a dozen locales should not assemble a dozen vendors. If voice cloning is central to what you are building, that is a capability they have made their own and we do not offer. And if you already run on their platform with little of your output in our three languages, the cost of moving is real. Our case is a product that speaks English, Thai, or Mandarin, where Thai sentences carry English words and the rate matters at volume.

Questions

Does ElevenLabs support Thai?

Yes. ElevenLabs' models documentation lists Thai for Eleven v4, Eleven v4 Turbo, Eleven v3, and Eleven v3 Conversational. Eleven Multilingual v2 and Eleven Flash v2.5 do not list it.

Sources:ElevenLabs models(opens in a new tab)

Can we run both?

Yes, and plenty of teams should. Route English, Thai, or Mandarin through us and keep everything else where it is; nothing about either API prevents it.

Do you publish quality comparisons?

No. We publish no accuracy or quality figures of our own, for our model or anyone else's, and will not until the lab publishes its evaluations. Listen to both on your own text and decide from that.

How current are the prices on this page?

They come from each vendor's published list rate, recorded in one place and re-checked on a fixed schedule. The month of the last check is printed at the top of the page.

Do you offer voice cloning?

No. Catalog voices are published with stable ids, and a voice of your own is one you design from a written description in the console. If cloning is a requirement, that requirement points elsewhere.

Other comparisons

Start with free credits

Start free