Switchboard
Conversational speech
Unscripted two-party telephone speech. WER % results.
Audio intelligence
Enable code-switching and transcribe any language mix in one request. Gladia detects the active language automatically as people speak, labels every word with its language code across 100+ languages, with no separate detection step to maintain.
Test code-switching on your own multilingual recordings. No expiry, no credit card.
Real-time transcripts that keep up, even when speakers switch languages mid-sentence.
Real conversations don’t stay in one language. Speakers switch mid-call, even mid-sentence, and single-language models garble those moments or drop them entirely. Gladia detects every switch automatically across 100+ languages, so each part is transcribed in the language it was spoken.
Compare Solaria-1 with leading speech-to-text APIs on the audio your product actually handles: spontaneous phone calls and overlapping speakers.
Code-switching works with both async and real-time transcription. There’s no language routing to configure and no second model to host.
Upload a file, pass a URL, or open a real-time WebSocket session.
Set "code_switching": true. Optionally list the languages you expect, which narrows detection and improves accuracy.
Each utterance and word comes back with its language code, ready to route to translation, NLP, or display.
Use cases
Skip the language configuration. Transcribe calls and meetings in any mix of languages, and keep summaries, sentiment, search, and speaker labels accurate, even when speakers switch mid-sentence.
Capture global team meetings where people switch languages, so notes and action items stay complete.
Learn more →Transcribe multilingual customer calls accurately, so QA scores and sentiment hold up in every language.
Learn more →Understand callers who mix languages in real time, without adding latency to every turn.
Learn more →Caption multilingual interviews and podcasts accurately, even when guests switch languages mid-sentence.
Learn more →
Code-switching and all 100+ languages are included in every paid plan at no extra cost. Your rate stays the same whether a call is in one language or three.
Flexible pay-as-you-go for moderate audio volumes. Get started immediately.
Async at $0.61/hr
Real-time at $0.75/hr
* 50€ in free credits
Lower unit pricing for fast-growing teams. Commit upfront to unlock savings.
Async as low as $0.20/hr
Real-time as low as $0.25/hr
* 67% less than Starter
Annual plan with custom models, fine-tuning, debundled pricing, and more.
Custom
Start free with 50€ in credits, or book a demo to test code-switching on your own audio.
