Request Demo or Early Access to our end-to-end speech-to-text API

By integrating our new API, you will gain access to features including:

  • Real-time transcription and speaker diarization in 50+ languages
  • Automatic language detection and switching for multilingual content
  • Identification of speakers through voiceprint technology
  • Silence detection with audio silence not billed, ensuring fair pricing
  • Transcribe any file, regardless of format, size or duration
  • Support for unlimited transcription concurrency
Request access
A developer holding a tablet in an open-plan office, over a Python snippet posting a recording to the Vatis Tech transcription API

Demo

See for yourself

Watch a quick demo of our new end-to-end multilingual speech-to-text technology.

Features

Everything the end-to-end API does

Streaming and pre-recorded, in one API. The rows marked with an asterisk are on the way; everything else is live today.
Capabilities of the Vatis Tech end-to-end speech-to-text API, by group
CapabilityAvailability
Transcription
Supported languages50+
Batch transcriptionIncluded
Live transcriptionIncluded
Automatic language detectionAutomatically detects the spoken language at the start of the audio and applies the correct transcription model.Included
Language code switchingAutomatically switch between different language models during the transcription process.Included
Batch speaker diarizationIncluded
Live speaker diarizationIncluded
Identify speaker by voiceprintAutomatically identifies and labels speakers from a voice sample.Included
Silence detectionIdentifies and marks quiet segments in audio.Included
Transcribe by a specific rangeTranscribes specified sections of audio.Included
Speech intelligence*
Batch audio intelligenceExtract insights, summarize speech, identify spoken topics, from pre-recorded audio files.Included
Live audio intelligenceReal-time insights from audio streams, summarize speech, identify spoken topics.Included
Translation*
Supported languages30+
Batch translationAutomatically translate the transcript of pre-recorded files into multiple languages.Included
Live translationAutomatically translate real-time transcription into multiple languages.Included
Input specs
Batch transcription supported formatsAny
Live transcription supported formatsAny
Audio lengthAny
File size**Any
ConcurrencyUnlimited
Output specs
Export formats*JSON, TXT, PDF, SRT, DOCX5
Infrastructure specs
Separate ingress and egress clientsManage incoming and outgoing data independently.Included
Multiple egress clientsMultiple clients can consume from the same real-time transcription stream.Included
Egress resumingAllows for the continuation of outgoing data transmissions after interruptions.Included
Ingress resuming*Enables the continuation of incoming data transmissions after interruptions.Included
Zero retention policyNo data is stored after processing.Included
Business model
Remove audio silence from billing usageExcludes silent audio segments from billing.Included
Price per hour for pre-recorded or streamingHourly pricing for pre-recorded or streaming content.Included
Price per channel for media monitoringPer monitored channel pricing for media monitoring companies.Included
Price per seat for contact centre solutions providersIncluded
  • *Will be available soon
  • **Limited at 3GB for mp4 and mov due to the lack of streaming compatibility
Euronews Romania

In our quest for a state-of-the-art transcription solution, Vatis Tech emerged as the clear choice, thanks to its solid accuracy, ease of use, and real-time capabilities.

After an extensive review and comparison of multiple speech-to-text solutions, including considerations for an in-house developed system, Vatis emerged as the superior solution.

Adrian VintilaHead of Broadcast IT, Euronews Romania

Early access

Request a demo or early access

Add your details and we will be in touch to walk you through the API and get you onto the early-access programme.

Loading the form… If it does not appear, email contact@vatis.tech instead.

For engineers who read the docs before the marketing page

Read the documentation, try it free, tell us how it goes.