Deepgram Flux (realtime STT)
unofficialrealtime
Deepgram Flux (realtime STT) specifications
| Context window | — |
|---|---|
| Max output tokens | — |
| Release date | — |
| Input | — |
| Output | — |
Deepgram Flux (realtime STT) pricing in rubles
You pay for what you use, per token and per unit. The final request cost is returned in the API response.
| Type | Price |
|---|---|
| Transcription | 1 ₽ per minute |
Prices are in rubles and may change with exchange rates.
What Deepgram Flux (realtime STT) can do
- Streaming responses
- Real-time transcription
Deepgram Flux (realtime STT) API: connection and code examples
The service exposes a single OpenAI-compatible API. Set our base_url and the key from your dashboard.
API endpoints
GET /v1/realtimeprimary | Realtime (WebSocket) Bidirektionaler Audio-Stream in Echtzeit über WebSocket. |
Supported request parameters
| Parameter | Default | Allowed values | Description |
|---|---|---|---|
encoding | linear16 | — | Audio-Stream-Codec für Echtzeitsitzungen. |
sample_rate | 16000 | — | Abtastrate des Audios in Hertz. |
language | — | — | Materialsprache: Prompt für das Modell, in welcher Sprache gesprochen oder der Text geschrieben ist. Beschleunigt und präzisiert in der Regel die Verarbeitung. |