# AI Transcription API | Transcribe Fast, Accurately, and in 55+ Languages

Source: https://www.speechmatics.com/product/transcription

Our Speech Recognition Transcription API allows businesses to transcribe audio & video into text accurately — in batch or real-time. 55+ languages ready now!

## AI transcription API, built for real-world performance.

Built for developers, trusted by enterprises—our AI transcription API combines low-latency with high-accuracy output, delivered on-prem or the cloud.

- [Book a Demo](https://www.speechmatics.com/speak-to-sales)
- [Get Started](https://portal.speechmatics.com/signup/)

Trusted by

- [AI Media](https://www.speechmatics.com/product/case-studies/transforming-live-captioning-how-ai-media-are-advancing-real-time) — Delivering 120X more with voice AI
- Pipecat
- mediQuo
- [LiveKit](https://www.speechmatics.com/company/articles-and-news/build-ai-agents-that-understand-who-said-what-livekit) — Enabling 100,000+ developers with leading speech recognition
- [Adobe](https://www.speechmatics.com/company/articles-and-news/adobe-and-speechmatics-deliver-cloud-grade-speech-recognition-on-device-for-premiere) — Cloud-grade speech recognition on-device for Adobe Premiere
- NVidia Inception Program
- Vapi
- Humetrix
- [NCI](https://www.speechmatics.com/product/case-studies/nci) — Redefining real-time captioning
- Veritone
- [Media Track](https://www.speechmatics.com/product/case-studies/media-track-enhances-global-media-monitoring-with-speechmatics) — Delivering a 20% leap in accuracy improvements
- Content Guru
- Docbuddy
- Cekura
- Jambonz
- Zylinc
- [Prosodica](https://www.speechmatics.com/product/case-studies/vail-systems-prosodica) — Driving better conversations at scale
- Ubisoft
- Edvak
- ACA Group
- Stenoly
- Nabla
- Speech Intelligence - 3playmedia
- Vodex
- [Stenograph](https://www.speechmatics.com/company/articles-and-news/stenograph-and-speechmatics-announce-industry-first-on-device-integration-for-catalyst-vp) — Industry-first on-device speech recognition for CATalyst VP

## Why developers choose our AI transcription API

### Accurate — AI transcripts you can trust

Trusted by enterprises & developers worldwide, our models deliver 90%+ accuracy across real-world use cases.

- [Get Started](https://portal.speechmatics.com/signup/)

### Low-latency — <500ms latency

Precise, low-latency transcription across 55+ languages, delivered before your media even ends.

- [Latency](https://docs.speechmatics.com/speech-to-text/realtime/output#partial-transcripts)

### Integration — Quick, flexible deployment

On-prem? Cloud? On-device? However you want it, we can provide it through our GPU infrastructure.

- [Deployment options](https://docs.speechmatics.com/deployments)

## Accuracy — Hitting the mark with pinpoint accuracy

Best in class ASR

We outperform the biggest companies in the world across the [languages we support](https://www.speechmatics.com/product/translation#against-competition).

Our inclusive ASR works **regardless of the accent or dialect**, even in challenging, noisy environments.

## Discover our AI transcription capabilities

Delivering for multilingual, multicultural, and multinational businesses.

- [View all features](https://docs.speechmatics.com/features)

## Global reach — 55+ languages

Supporting transcription in [55+ languages](https://www.speechmatics.com/product/translation) with automatic language detection.

- [Explore the docs](https://docs.speechmatics.com/speech-to-text/languages)

## Punctuation and numerals — Smart formatting

Correctly [formatted numbers](https://docs.speechmatics.com/features/entities), dates, and currencies, as well as language-specific capitalization (e.g. "one thousand" to "1000").

- [Smart formatting](https://docs.speechmatics.com/features/punctuation-settings)

## Customization — Custom Dictionary

Boost accuracy for proper nouns, acronyms, or industry-specific terms by providing a list of custom words.

- [Custom dictionary](https://docs.speechmatics.com/features/custom-dictionary)

## AI transcription — Real-time & pre-recorded

Live or pre-recorded, our models deliver unmatched accuracy and speed—outperforming every other solution.

- [File and real-time transcription](https://docs.speechmatics.com/speech-to-text/batch/quickstart)

## Multi-speakers — Diarization

Diarization identifies and labels multiple speakers in complex conversations, even in real-time environments.

- [diarization link](https://docs.speechmatics.com/features/diarization)

## Disfluencies — Filler words

Capture interruptions like “huh” and “hmm” to reflect more natural, conversational speech.

- [Disfluencies](https://docs.speechmatics.com/features/word-tagging)

## Need speed? Prefer accuracy?

Choose your operating point and get exactly what you need. We offer three proprietary transcription models available to all customers, including [Melia](https://www.speechmatics.com/company/articles-and-news/introducing-melia-multilingual-speech-to-text-model):

### Standard

##### Great for users and generating transcripts where speed is a priority, with accuracy trade-offs as a result.

- [Pricing](https://www.speechmatics.com/pricing)

### Enhanced

##### When unbeatable accuracy is a must-have, our Enhanced model provides best-in-class accuracy across all of our languages.

- [Pricing](https://www.speechmatics.com/pricing)

## _“Working with Speechmatics enables us to seamlessly provide our customers with_ **_quality, automated speech analytics_** _as part of our solution."_

[Mariano Tan](https://www.linkedin.com/in/marianotan/), President & CEO, Prosodica

- [Prosodica's Story](https://www.speechmatics.com/product/case-studies/vail-systems-prosodica)

## _"We're delighted to work with Speechmatics to drive our live and batch captioning – they continue to be_ **_ahead of the pack_** _for all key quality metrics."_

[Tom Wootton](https://www.linkedin.com/in/tom-wootton-05b282139/), Product Leader, Red Bee

- [Red Bee Media's Story](https://www.speechmatics.com/product/case-studies/red-bee-media)

## _"They consistently outperform other vendors for_ **_word error rate and punctuation_** _- playing a pivotal role in the development of our workspace."_

[Maarten Verwaest](https://www.linkedin.com/in/maartenverwaest/), CRO, Limecraft

- [Limecraft Case Study](https://www.speechmatics.com/product/case-studies/limecraft-help-broadcasters-expand-audiences-using-ai)

## **Try It Now. For Free. Without Code.**

The BEST way to view Speechmatics' accuracy is to see for yourself, on your media. 
Head to the portal and get a free account today.

- [Try It Now. For Free. Without Code.](https://portal.speechmatics.com/signup/)

## Resources

### Languages — Speechmatics Medical Model launches in Spanish

- [Speechmatics launches medical model - carousel](https://www.speechmatics.com/company/articles-and-news/speechmatics-medical-model-launches-in-spanish)

Joining French, Dutch, Finnish and English for global clinical transcription - accurate, hallucination-free, and accent-independent.

- Speechmatics — Editorial Team

### Voice Agents — Vapi and Speechmatics: Build agents that understand every voice

- [Vapi and Speechmatics build better agents](https://www.speechmatics.com/company/articles-and-news/vapi-and-speechmatics-build-agents-that-understand-every-voice)

Ship Voice AI agents that stay readable in real time, even in noisy, multi-speaker calls.

- Speechmatics — Editorial Team

### Text-to-Speech — Why we built our low-latency Text-to-Speech

- [Why we built our low-latency TTS](https://www.speechmatics.com/company/articles-and-news/why-we-built-our-low-latency-text-to-speech)

Most TTS sounds great in demos but breaks in real conversations. We built ours for sub-150ms latency, natural voices, and global scale.

- Stuart Wood — Product Manager

### Medical — The ultimate guide to healthcare speech recognition

- [What is AI medical transcription?](https://www.speechmatics.com/company/articles-and-news/what-is-ai-medical-transcription-the-ultimate-guide-to-healthcare-speech)

Reducing documentation time, easing physician burnout, and improving patient care and efficiency with Voice AI.

- Blair Robertson — Account Executive

### On-Prem — The return of on-premise: Why enterprise AI's head is no longer in the cloud

- [The return of on-premise: Why enterprise AI's head is no longer in the cloud](https://www.speechmatics.com/company/articles-and-news/the-return-of-on-prem-why-enterprise-is-no-longer-in-the-cloud)

As regulations rise and cloud costs spiral, enterprises are bringing AI home—with better outcomes.

- Brad Phipps — Director, SaaS & Infrastructure

### Voice Agents — Introducing real-time, speaker-aware Voice Agents with LiveKit + Speechmatics

- [Introducing real-time, speaker-aware Voice Agents with LiveKit + Speechmatics](https://www.speechmatics.com/company/articles-and-news/build-ai-agents-that-understand-who-said-what-livekit)

Speechmatics brings speaker diarization to LiveKit agents - enabling them to understand not just _what_ was said, but _who_ said it.

- Anthony Perera — Product Marketing Manager