Intelligent real-time and asynchronous transcription for video and audio

Overview enables accurate speech-to-text capabilities for variety of use cases, across all conversation channels. Transcribe conversations in real-time though WebSockets and a variety of streaming protocols, or asynchronously from recorded audio, video, and text files.

Best in class accuracy

Breakthrough ‘Unified Modeling’ approach resulting in lower word error rate (WER) and word information loss (WIL) compared to other cloud speech recognition offerings.

Real-time and asynchronous

Transcribe conversations in real-time through WebSockets and a variety of streaming protocols, or asynchronously from recorded audio and video files

High quality transcription for your data through custom vocabulary

Add custom vocabulary or customize the speech model to further increase transcription quality by with your own industry terms, keywords and phrases

Multi-streaming connections with speaker separation

Supports unlimited streaming audio connections in a single session and can identify distinct speakers and predict which utterances belong to whom.

Supports multiple languages, accents and dialects

Over 20 languages and accent variations are currently supported.

Paragraph formatting and punctuation

Export your transcription as SRT or markdown for higher readability and directly plug into video players for closed captions

Use Cases

Real-time captioning

Add subtitles to live video conferencing or webinars for seamless collaboration or add captioning to a customer care call for agent assistance.

Search and Accessibility

Process archived audio and video files to create easy to read, searchable conversations, unlocking data hidden away in thousands of hours of files.

Conversation Analysis

Monitor conversations with customers. In conjunction with Symbl’s other powerful APIs unlock insights and speaker analytics and improve the quality of each interaction.


Processing asynchronous conversations

Submit recorded or saved conversations in video and audio formats through the Async API.

Processing real-time conversations

Get an active WebSocket connection to the Streaming API for live conversation transcription and intelligence.

Transcribe webinars & conferences

Using Subscribe API, you can transcribe conversations where there are only a handful of speakers and most participants are only listening in.