Transcribe live and recorded audio via Deepgram with speaker labels.
npx skills add https://github.com/mattwoodco/skills --skill transcription
Please help me install this Agent Skill. Skill: transcription Source: https://github.com/mattwoodco/skills/tree/main/skills/transcription Command: npx skills add https://github.com/mattwoodco/skills --skill transcription
This Skill automates the process of converting spoken audio into written text, enabling real-time captions and searchable post-meeting transcripts.
Use the transcription skill to set up live speech-to-text for the current video room.
High-intent search queries and answers about installing and using this skill.
Real-time speech-to-text for a video room is provided through live streaming transcription using Deepgram's API. It overlays live captions directly within your video conferencing application during active meetings.
Batch transcription processes pre-recorded audio files to generate full post-meeting transcripts. It includes speaker diarization to identify and label different speakers, producing searchable text outputs.
You need PostgreSQL with Drizzle ORM setup to persist transcripts. The database integration stores both real-time and batch speech-to-text outputs, enabling searchable post-meeting transcript retrieval.
Setting up Deepgram transcription requires a Deepgram API key, video room integration, database setup, and environment configuration. These prerequisites enable live streaming speech-to-text capabilities for your application.
Deepgram supports batch processing of pre-recorded audio files for recorded lectures. It converts spoken audio into searchable written text transcripts, suitable for generating accessible post-lecture documentation.
Deepgram real-time transcription requires stable video room integration and proper environment configuration. Without a PostgreSQL database via Drizzle ORM, live speech-to-text transcripts cannot be persisted for later retrieval.
aberrantCodeCreate Node.js APIs with Supabase authentication, Drizzle ORM, and Zod validation.
CaneproTranscribe audio with optional speaker diarization and known-speaker references.
FutureAtomsIntegrate LiveKit React Native SDK for real-time audio and video communication.
Shakudo-ioConvert spoken language into searchable text via the Nova-3 model.
