Paxa Labs and Gladia for multilingual speech to text

A transcription API for Thai and other languages next to Gladia, across code switching, language support, how the transcript comes back, speakers, vocabulary, subtitles, and live turns.

Gladia's Solaria-1 lists Thai and Chinese and switches between languages when code switching is on, and its guide recommends a short list of the languages to expect. Its Solaria-3 covers English, French, German, Italian, and Spanish. Paxa STT Lite transcribes Thai, English, Mandarin, and other languages. A Thai sentence that turns into English is written as it was spoken, with no setting. Paxa STT Lite is a Research preview. Requests bill at the published rate, and its transcript quality and options may change between releases.

Sources:Gladia supported languages(opens in a new tab)Gladia code switching(opens in a new tab)

Rates and claims last checked October 2026

Start free See full pricing

What differs

What differs
Dimension	Paxa Labs	Gladia(opens in a new tab)
Thai with English inside it	A Thai sentence that switches into English is written as spoken, with no setting. Nothing is translated.	Code switching on Solaria-1, set with language_config.code_switching. Their guide recommends a short list of expected languages and returns the detected language with each utterance and word.
Languages	Thai, English, Mandarin, and other languages. Paxa STT Lite writes speech in the language spoken and detects that language itself. A BCP 47 hint corrects a recording it misreads.	Thai and Chinese on Solaria-1, which covers every language on their table. Solaria-3 covers English, French, German, Italian, and Spanish.
Getting the transcript	One request with model paxa-stt-lite-v1-preview returns the transcript in its response, for recordings up to 120 minutes and 95 MiB.	Upload the file, create a job with the returned URL, then poll, or wait for a webhook or callback. Up to 135 minutes and 1000 MB per request, or 4 hours 15 minutes on enterprise plans.
Speaker turns	Speaker turns numbered from 0 in order of first appearance, on recordings up to 120 minutes, at no extra charge.	Diarization on pre-recorded audio, numbered from 0 in order of first appearance, with speaker-count hints. Their pricing page lists it as async only.
Vocabulary	A vocabulary of up to 50 terms, in Thai or any other language, at no extra charge.	A custom vocabulary on pre-recorded and live audio, each term with optional spelled-out pronunciations, an intensity, and a language. Their guide warns that long lists increase false positives.
Subtitles	An SRT or WebVTT file in the same response. Every line break falls on a word boundary, Thai included.	SRT and VTT from a pre-recorded transcript, with limits on characters per row and rows per caption, and a compliance style.
Live transcription	Paxa STT Lite Realtime streams over a WebSocket, reports when the user starts and stops talking, and sends one final transcript per turn.	Solaria-1 over a WebSocket, with speech begin and end events. An utterance closes on an endpointing duration the client sets, 0.05 seconds by default, and a session lasts up to 3 hours.

Sources:Gladia code switching(opens in a new tab)Gladia supported languages(opens in a new tab)Gladia pre-recorded transcription(opens in a new tab)Gladia supported files and duration(opens in a new tab)Gladia speaker diarization(opens in a new tab)Gladia pricing(opens in a new tab)Gladia custom vocabulary(opens in a new tab)Gladia subtitles(opens in a new tab)Gladia live transcription(opens in a new tab)Gladia live session reference(opens in a new tab)

Paxa STT Lite, per hour of audio: $0.50
Gladia: $0.18-0.30
Their rate, as a multiple: 0.36-0.61×

Gladia's range runs from Solaria-3, which covers five languages without Thai, to Solaria-1, at its self-serve async rate for a month's first 100 hours. Gladia prices in euros, converted at the European Central Bank reference rate of 1.1224 US dollars to the euro. Every rate is per hour of recorded audio, from the vendor's cheapest pay-as-you-go model to its most expensive. Per-minute and per-second rates are converted to that unit.

What it costs at your volume

What it costs at your volume
Audio, per month	Paxa Labs	Gladia	Difference
10 hours	Paxa Labs$5	Gladia$1.80-3.03	Difference$3.20 more
100 hours	Paxa Labs$50	Gladia$17.96-30.30	Difference$32.04 more
1,000 hours	Paxa Labs$500	Gladia$179.58-303.05	Difference$320.42 more
10,000 hours	Paxa Labs$5,000	Gladia$1,795.84-3,030.48	Difference$3,204.16 more

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 Gladia is the better choice

Ours is a Research preview, and its output and options may change between releases. A team that needs a settled model today has a reason to stay. Gladia takes recordings up to 135 minutes, longer on enterprise plans, and sells summarization, translation, and named entity recognition as hourly add-ons. Its Enterprise plan offers custom models and fine-tuning, and we offer neither. Choose us when Thai speakers switch into English and one request should return the transcript.

Questions

Does Gladia transcribe Thai?

Yes, on Solaria-1, and their language table marks code switching for Thai. Solaria-3, the lower end of the range here, covers five languages, and Thai is not among them.

What does Research preview mean for Paxa STT Lite?

Requests bill at the published rate. Transcript quality and the option set may change between releases, and the changelog on our speech to text page records each change.

Do you publish accuracy or latency comparisons?

No. We publish no accuracy, quality, or latency figures of our own, for our model or anyone else's. Transcribe the same recording on both and read the two transcripts.

How current are the prices here?

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 at the top of the page.

Other comparisons

Start with free credits

Start free

Product

Speech-to-Text

Use cases