Overview

Welcome to SubQ - a fast, accurate speech-to-text API that is fully compatible with the Deepgram API.

Why SubQ?

SubQ gives you production-ready speech-to-text with minimal integration effort:

• Deepgram-compatible: Use existing Deepgram SDKs by changing two lines of code (API key + base URL)
• Pre-recorded & streaming: Batch transcription via REST or real-time transcription via WebSocket
• Mobile SDKs: First-party Swift (iOS) and Kotlin (Android) SDKs
• Rich feature set: Speaker diarization, smart formatting, language detection, PII redaction, and more

How it works

SubQ exposes a /v1/listen endpoint that accepts audio and returns transcription results. You can send audio in two ways:

1. Pre-recorded: POST an audio file or URL and receive the full transcript in the response
2. Streaming: Open a WebSocket connection and send audio chunks for real-time, low-latency transcription

Integration paths

Use existing Deepgram SDKs

If you already use Deepgram, switch to SubQ by updating your API key and base URL. All existing Deepgram SDKs (Python, Node.js, Go, Rust, .NET) work out of the box.

Use raw HTTP

Call the API directly with any HTTP client: no SDK required.

Next steps