skills.rest
framerslab
framerslabOFFICIAL
Explore all skills by this author →
Website𝕏@framerslabGitHub

streaming-stt-deepgram

Transcribe live speech via Deepgram's WebSocket API with speaker diarization.

1|Updated Feb 8, 2026
Search Tags:#real-time#streaming#speech-to-text#deepgram#diarization
One-click install
npx skills add https://github.com/framerslab/agentos-skills --skill streaming-stt-deepgram
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: streaming-stt-deepgram
Source: https://github.com/framerslab/agentos-skills/tree/main/registry/curated/streaming-stt-deepgram
Command: npx skills add https://github.com/framerslab/agentos-skills --skill streaming-stt-deepgram

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides real-time speech-to-text transcription with low latency, using Deepgram's WebSocket API, suitable for live voice input during conversations and speaker diarization.

Core Features & Use Cases

  • Real-time Transcription: Offers sub-300 ms latency transcription.
  • Speaker Diarization: Identifies and labels speakers in multi-person conversations.
  • Auto-Reconnect: Automatically reconnects to WebSocket API on drops.
  • Use Case: Ideal for live voice sessions where immediate transcription is needed, such as customer service interactions or conference calls.

Quick Start

Start a live voice session using Deepgram for transcription by setting the environment variable DEEPGRAM_API_KEY.

Frequently Asked Questions about streaming-stt-deepgram

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get real-time speech-to-text transcription with low latency?▼

Real-time speech-to-text transcription with sub-300 ms latency is achieved by streaming live audio through Deepgram's WebSocket API, returning immediate text results during active voice sessions.

How does speaker diarization work for live multi-person conversations?▼

Speaker diarization identifies and labels individual speakers in real-time during multi-person conversations, assigning transcription text to specific participants as audio streams through the WebSocket connection.

What do I need to set up before using Deepgram's WebSocket API for transcription?▼

You need a valid Deepgram API key set as an environment variable and an active WebSocket connection to start streaming live voice audio for transcription.

Does this handle WebSocket connection drops during live transcription?▼

Yes, an auto-reconnect mechanism automatically re-establishes the WebSocket connection if drops occur, ensuring continuous real-time transcription without manual intervention.

When should I use streaming speech-to-text over batch transcription?▼

Streaming speech-to-text is ideal for live voice sessions requiring immediate transcription, such as customer service interactions or conference calls, where low latency and speaker diarization are critical.

Can I use this for conference call transcription with multiple speakers?▼

Yes, conference call transcription is supported through real-time streaming via Deepgram's WebSocket API, utilizing speaker diarization to label multiple participants with sub-300 ms latency.

Related Skills

deepgramdeepgram
456

deepgram-python-speech-to-text

Transcribe audio files and streams using the Deepgram Python SDK.

Official
Intermediate
mehdicmehdic

quality-dashboard

Aggregate security, coverage, and lint JSON reports into a unified health score.

Community
Intermediate
jacob-splitjacob-split

minutes-record

Record meetings and convert them into searchable markdown transcripts.

Community
Intermediate
aztr0nutzsaztr0nutzs
1

Deepgram Speech-to-Text SDK

Transcribe live audio streams and pre-recorded files via Deepgram SDK.

Community
Intermediate