vsref
Gladia logo

Gladia Review

EU-based real-time and batch STT API built on the Solaria models

Among the 43 speech-to-text tools we track, Gladia has the 4th-widest language coverage - a fit for multilingual and localization projects.

See pricing

Facts verified Jul 20, 2026Try Gladia →

If you sign up through links on this page, vsref may earn a commission; programs exist on both sides of most comparisons, and commissions never change verdicts. How we make money

What we know about Gladia

Gladia sits in the speech-to-text apis category, where it is eU-based real-time and batch STT API built on the Solaria models. We keep this Gladia profile grounded in primary sources, each fact dated to when we last confirmed it.

Gladia does not publish a public per-minute rate we have been able to verify, so the price figure here stays blank until we can confirm one. We would rather show nothing than a guessed number; if you have a current Gladia quote, it is the fastest way for us to close that gap.

On capabilities, Gladia covers language auto-detection, word-level timestamps, custom vocabulary / keyterm boosting, entity detection, sentiment analysis, and summarization endpoint. Each of those is verified against Gladia's own docs or dashboard, not marketing copy.

For compliance, with Gladia: SOC 2 Type II is in place. If you are in a regulated space, confirm the current posture with Gladia before you commit, since these change plan by plan.

Among the 43 speech-to-text tools in our matrix, Gladia leads with the 4th-widest language coverage; the fact sheet below has the raw numbers behind that placement.

In total we track 28 verified facts for Gladia today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the Gladia fact sheet below.

Reviewed by vsref Editorialfacts verified Jul 20, 2026Methodology →

Gladia pricing

Published rates: batch $0.0102/min · streaming $0.0125/min, verified Jul 20, 2026 (source).

Gladia cost at monthly volume tiers
Monthly volumeBatch billStreaming bill
1K min/moSide project$10.17$12.50
10K min/moProduction app$101.67$125
100K min/moCall-center scale$1,016.70$1,250

Sticker rates only; diarization and PII redaction add-ons price in the stack builder. How we compute costs →

Fact sheet

Pricing
Pricing facts
Batch price per audio minute0.01 $/audio-minJul 20
Streaming price per audio minute0.013 $/audio-minJul 20
Pricing modelusageJul 20
Free tier quota10h free monthlyJul 20
Published volume discountsGrowth: async as low as $0.20/hr, streaming $0.25/hrJul 20
Concurrency on base plan25 async / 30 real-time concurrent (paid)Jul 20
Capabilities
Capabilities facts
WER (third-party benchmark)3.23 % WERJul 20
WER (vendor-claimed)~9.6 % WERJul 20
Streaming latency (vendor-claimed)~300 msJul 20
Languages supported101 languagesJul 20
Language auto-detection✓ YesJul 20
Speaker diarization✓ IncludedJul 20
PII redaction✓ IncludedJul 20
Word-level timestamps✓ YesJul 20
Custom vocabulary / keyterm boosting✓ YesJul 20
Entity detection✓ YesJul 20
Sentiment analysis✓ YesJul 20
Summarization endpoint✓ YesJul 20
Speech translation✓ YesJul 20
Priced audio-intelligence add-onsSummarization, translation, sentiment, NER, PII redactionJul 20
Compliance & trust
Compliance & trust facts
HIPAA BAA available✓ YesJul 20
SOC 2 Type II✓ YesJul 20
GDPR / EU data residency✓ YesJul 20
Self-host / on-prem option✓ YesJul 20
Build experience
Build experience facts
Official SDKsPython, JS/TSJul 20
Max file size / duration1000 MB; 135 min per request (4h15 Enterprise)Jul 20
Supported audio formatsaacJul 20
Websocket streaming API✓ YesJul 20

Considering a switch? Best Gladia alternatives →

Voice agents in this stack

The engine is one layer: a voice agent hears through its transcription engine, but orchestration, telephony, and turn-taking come from the agent platform. Compare the platforms builders pair Gladia with: Pipecat, OpenAI Realtime API, Twilio ConversationRelay, or the full voice-agent comparison →

Want it done for you?

Gladia gets you a transcript; the bot that joins the call, labels speakers, and writes the summary is still your build. If that is more pipeline than you want to own, AI meeting notetakers do the whole job end to end: Otter.ai, Fireflies.ai, Fathom, or the full notetaker comparison.

Gladia head-to-head

Gladia vs Deepgram →won 1 · lost 4 · tied 2
Call CenterslostDeveloperslostDictationtieMedicallostMeetingswonSelf-HostedtieVoice Agentslost
Gladia vs AssemblyAI →won 0 · lost 4 · tied 3
Call CenterslostDeveloperslostDictationtieMedicaltieMeetingslostSelf-HostedtieVoice Agentslost
Gladia vs OpenAI Whisper (API) →won 3 · lost 2 · tied 2
Call CenterstieDeveloperslostDictationtieMedicalwonMeetingswonSelf-HostedlostVoice Agentswon