Gladia
Gladia is an EU-based speech-to-text API built around the Solaria-1 model — sub-300 ms real-time streaming plus pre-recorded transcription, with 100+ languages, mid-utterance code-switching, and contractual EU data residency.
Drop your audio. Transcript in seconds. First transcript free, then $2 a file or $8 = 1,000 min
Solaria-1 is Gladia's universal STT model — one model id covers 100+ languages with automatic detection and mid-utterance code-switching, partial transcripts under ~100 ms, and built-in diarization, named-entity recognition, sentiment, and summarization. The platform exposes two surfaces: POST /v2/pre-recorded for async batch jobs and POST /v2/live → wss://api.gladia.io/v2/live?token=… for streaming, both keyed off the x-gladia-key header.
Best for voice agents, contact-center analytics, multilingual meeting bots, and any workload where EU data residency or unusual language coverage is the deciding factor. Pay-as-you-go on the Starter plan runs $0.61/hr async (~$0.0102/min) and $0.75/hr real-time (~$0.0125/min) with 10 free hours per month; the Growth tier discounts those to ~$0.20/hr async and ~$0.25/hr real-time with an upfront commit; Enterprise adds unlimited concurrency, zero retention, SLAs, and custom hosting. Last price check: 2026-05-10.
What it is
Gladia wraps Whisper-class models in a developer-friendly API with diarization, 99 languages, and competitive per-minute pricing. A reasonable alternative to self-hosting faster-whisper when you want someone else to operate the GPUs. Last price check: 2026-04-20.
Watch out for: Smaller ecosystem than AssemblyAI/Deepgram; HIPAA on enterprise tiers only.
Install / use
Where Gladia fits · 6 use-cases
Gladia's strengths cluster around multilingual accuracy, low-latency partials, and an EU compliance posture that most US-headquartered APIs can't match. Pick the card closest to your build — each links to the canonical docs section.
Initiate a live session with POST /v2/live, then stream PCM over the returned WebSocket. Partials arrive in roughly 100 ms, which is the latency budget conversational agents need before turn-taking feels broken. Audio-to-LLM lets a single request return both the transcript and a structured LLM response.
Sub-300 ms final · ~100 ms partials
Batch agent + customer recordings through POST /v2/pre-recorded with diarization on, optional speaker count, PII redaction, sentiment, and summarization in one call. SOC 2 Type II, HIPAA, and ISO 27001 are in scope on paid plans.
Single request returns transcript + speakers + summary
Solaria-1 covers 100+ languages under a single model id and handles mid-utterance switches without re-routing. Gladia's own benchmark calls out 42 languages that competing API vendors don't publish coverage for at all — useful when your inputs aren't English-first.
Detect + transcribe + code-switch in one pass
Send the episode URL or upload bytes, ask for diarization, paragraphs, SRT/VTT subtitles, and a summary in one POST. The audio-to-LLM feature can replace a separate summarization vendor for show-notes and chapter generation.
Publishable transcript + chapters in one call
Enterprise plans offer contractual EU-only data residency, zero retention, GDPR plus HIPAA plus SOC 2 Type II plus ISO 27001, and a no-training-on-customer-audio clause. This is the differentiator most US-headquartered APIs cannot match without a separate enterprise paper trail.
Zero retention available on request
Open the WebSocket returned from /v2/live and push PCM chunks; interim transcripts arrive with word-level timestamps for caption overlays, live-event accessibility, and broadcast workflows. A single live session is capped at three hours per the docs.
Word-level timestamps inline
Quickstart · pick a runtime
Three working ways to call Gladia. Export your key as GLADIA_API_KEY first — grab one from the Gladia console (10 free hours per month on Starter, no card required).
Two-step POST + poll against /v2/pre-recorded · transcribe any HTTPS audio URL with Solaria-1.
# pip install requests
import os, time, requests
API = "https://api.gladia.io/v2"
KEY = os.environ["GLADIA_API_KEY"]
HDR = {"x-gladia-key": KEY, "Content-Type": "application/json"}
# 1. submit
body = {
"audio_url": "https://files.gladia.io/example/audio-transcription/split_infinity.wav",
"diarization": True,
"subtitles": True,
"subtitles_config": {"formats": ["srt"]},
}
r = requests.post(f"{API}/pre-recorded", json=body, headers=HDR).json()
result_url = r["result_url"]
# 2. poll
while True:
res = requests.get(result_url, headers={"x-gladia-key": KEY}).json()
if res["status"] == "done":
print(res["result"]["transcription"]["full_transcript"])
break
if res["status"] == "error":
raise RuntimeError(res)
time.sleep(2)Plain HTTPS POST to /v2/pre-recorded · useful for shell pipelines and edge runtimes.
# submit a job
curl --request POST \
--url 'https://api.gladia.io/v2/pre-recorded' \
--header "x-gladia-key: $GLADIA_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
"audio_url": "https://files.gladia.io/example/audio-transcription/split_infinity.wav",
"diarization": true,
"subtitles": true,
"subtitles_config": {"formats": ["srt"]}
}'
# response includes { id, result_url }
# poll until status=done
curl --request GET \
--url "$RESULT_URL" \
--header "x-gladia-key: $GLADIA_API_KEY"Init the live session with POST /v2/live, then stream PCM to the returned wss:// URL.
// npm i ws node-fetch
import fetch from "node-fetch";
import WebSocket from "ws";
const KEY = process.env.GLADIA_API_KEY;
// 1. init session
const init = await fetch("https://api.gladia.io/v2/live", {
method: "POST",
headers: {
"x-gladia-key": KEY,
"Content-Type": "application/json",
},
body: JSON.stringify({
encoding: "wav/pcm",
sample_rate: 16000,
bit_depth: 16,
channels: 1,
}),
}).then((r) => r.json());
// 2. open the WebSocket
const ws = new WebSocket(init.url);
ws.on("open", () => {
// send 16 kHz / 16-bit PCM frames here
});
ws.on("message", (msg) => {
const evt = JSON.parse(msg.toString());
if (evt.type === "transcript" && evt.data?.is_final) {
console.log(evt.data.utterance.text);
}
});What people actually do with Gladia-style transcription
The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.
Features
| Speaker diarization | Yes |
| Word-level timestamps | Yes |
| Streaming / real-time | Yes |
| Languages supported | 99 |
| HIPAA eligible | No |
Links
- gladia.io ↗Product homepage — EU-based STT API positioning, Solaria-1 model, voice-agent and contact-center use cases.
- docs.gladia.io ↗Documentation root — quickstarts for pre-recorded, real-time, and audio-intelligence feature surfaces.
- API reference · pre-recorded ↗POST /v2/pre-recorded — request schema for diarization, subtitles, summarization, audio-to-LLM, plus the result-poll endpoint.
- API reference · live ↗POST /v2/live — init payload (encoding, sample rate, channels) and the wss URL pattern returned for the audio stream.
- Solaria-1 model page ↗Model card — 100+ language coverage, ~100 ms partials, accuracy claims on EN/ES/FR/IT benchmarks, code-switching.
- gladia.io/pricing ↗Current Starter, Growth, and Enterprise tiers — async, real-time, free-hour allowance, and zero-retention options.
- status.gladia.io ↗Live uptime for the Application, Pre-Recorded, and Real-Time components — subscribe via email or RSS.
- gladia.io/blog ↗Product blog — recent posts cover audio-to-LLM in a single POST, summarization, and Solaria-1 release notes.
Gladia vs Whipscribe
| Feature | Gladia | Whipscribe |
|---|---|---|
| Category | Transcription APIs | Transcription APIs |
| Pricing | Not verified | $8–$24 one-time packs (credits never expire) · $2 single unlock · free instant preview |
| Speaker diarization | Not verified | Yes |
| Word timestamps | Not verified | Yes |
| Streaming | Not verified | No |
| Languages | 99 | 99 |
| Platforms | API | Web, API, MCP |
Where this category is heading
From the vendor changelogs we track weekly — what changed in August 2026, and what it means if you are choosing now.
AssemblyAI moved summarisation onto an LLM this month; every vendor is racing to return action items, quotes and topics with the text rather than as an add-on.
Whipscribe today Every Whipscribe job already returns an insights payload — summary, key quotes, topics and speakers — from the same job id, at no extra charge.
Deepgram shipped self-hosted container images in August — the market is moving toward audio that stays inside a boundary the customer controls, because teams with customer calls or unreleased material are refusing shared model endpoints.
Whipscribe today Whipscribe runs on our own GPUs in a private, secured cloud. Audio is never forwarded to OpenAI or any third-party model.
The fastest-growing way to use a transcription API is not a form — it is Claude, Cursor or a workflow runner calling it mid-task through MCP.
Whipscribe today Whipscribe ships an MCP server: transcribe, search and summarise from an assistant without wiring anything.
AssemblyAI's 1.0 SDK unified async, realtime and sync; Deepgram's CLI went to 0.3. The unit of work is becoming the folder or the bucket, not the file.
Whipscribe today Submit with an Idempotency-Key and a batch_id, poll by job, retry safely. The S3 connector runs a whole prefix in one grant.
Deepgram added Afrikaans, Georgian and Armenian and improved a dozen more this month. Coverage is widening while quality still clusters around English and the large European languages.
Whipscribe today 99+ languages auto-detected. Ask for a language explicitly when you know it — auto-detect on a short or noisy clip is the most common cause of a wrong-language transcript.
Source: Deepgram and AssemblyAI changelogs, scanned 2026-08-24.
Alternatives to Gladia
Frequently asked about Gladia
What is Gladia?
Solaria-1 is Gladia's universal STT model — one model id covers 100+ languages with automatic detection and mid-utterance code-switching, partial transcripts under ~100 ms, and built-in diarization, named-entity recognition, sentiment, and summarization. The platform exposes two surfaces: POST /v2/pre-recorded for async batch jobs and POST /v2/live → wss://api.gladia.io/v2/live?token=… for streaming, both keyed off the x-gladia-key header.
How much does Gladia cost?
Gladia is a paid product — published pricing: from $0.0102/min. Pricing changes; verify on the vendor's own page before budgeting.
What platforms does Gladia support?
Gladia is an API — you call it from whatever you build, on any platform with an HTTP client.
How do I get started with Gladia?
Gladia is used through its API: get a key from the vendor, then call it from your own code. There is no desktop app to install.
How many languages does Gladia support?
Gladia lists 99 languages. That is the Whisper-family multilingual set, so quality varies by language — English and the large European languages are strongest.
What are the limitations of Gladia?
Smaller ecosystem than AssemblyAI/Deepgram; HIPAA on enterprise tiers only.
Who is Gladia best for?
Teams who like the Whisper model family but don't want to run GPUs.
What should I know before choosing Gladia?
Best for voice agents, contact-center analytics, multilingual meeting bots, and any workload where EU data residency or unusual language coverage is the deciding factor. Pay-as-you-go on the Starter plan runs $0.61/hr async (~$0.0102/min) and $0.75/hr real-time (~$0.0125/min) with 10 free hours per month; the Growth tier discounts those to ~$0.20/hr async and ~$0.25/hr real-time with an upfront commit; Enterprise adds unlimited concurrency, zero retention, SLAs, and custom hosting. Last price check: 2026-05-10.
Is Gladia open source?
No. Gladia is a proprietary API. If you need source you can read and run yourself, the open-source tools in this directory are the place to look.
What kind of tool is Gladia?
In this directory Gladia is filed under commercial api as a API.
What are the alternatives to Gladia?
There is a side-by-side page at /tools/gladia-alternatives comparing Gladia with the closest tools in the same category on price, platform and features.
Are there setup recipes for Gladia?
Yes — this page carries 3 tested setups: Python · pre-recorded URL; cURL · no SDK; Node · real-time WebSocket.
Can I automate Gladia-style transcription without running it myself?
Yes. If what you want is transcripts rather than the tool itself, Whipscribe does the same job as a hosted API: submit a file or URL, poll a job id, pull the result as txt, json, srt, vtt or docx. $0.008 a minute — about $0.48 an audio hour — bought as credits that never expire. The automation recipes on this site show it wired to Drive folders, Zoom recordings, S3 buckets and no-code platforms.
How is Whipscribe different from Gladia?
Gladia is another vendor's product; Whipscribe is a per-minute transcription service. We run the models on our own GPUs in a private cloud — audio is never forwarded to OpenAI or any third-party model — and every job returns five formats plus an AI summary, quotes and topics. Your first transcript in the web app is free at any length, no card.
What does transcription cost if I use Whipscribe instead of Gladia?
$0.008 a minute — about $0.48 an audio hour — bought as credits that never expire. There is no subscription and no monthly minimum: a 3-hour recording is about $1.44, a thousand-hour archive is about $480 once. Your first transcript in the web app is free at any length, no card, so you can check accuracy on your own audio before paying anything.
Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.
Explore
All transcription tools · Audio technology hub · Transcribe any platform · Audio & video formats · How-to guides · Glossary · Playbooks · Apps · Broadcast & radio · Podcast transcripts · Use cases · Blog · Transcription API · Integrations · Automations · For your industry