What problem does it solve? Converting audio recordings into readable transcripts with correctly attributed speakers is tedious manual work. This Skill automates transcription of audio files or URLs via the Deepgram API, identifies distinct speakers, and lets you assign real names to each speaker for a clean, readable transcript. ## Core Features & Use Cases - Audio Transcription with Diarization: Transcribes local audio files or remote URLs using Deepgram's nova-3 model with speaker separation and smart formatting. - Interactive Speaker Naming: Presents the first turns of each speaker so users can identify who's who, then maps speaker IDs to real names or descriptions. - Multilingual Support & Caching: Accepts a BCP-47 language tag (e.g., Spanish) and caches raw API responses so formatting never triggers a second paid API call. - Use Case: Transcribe a recorded podcast episode from a URL, identify the host and guest from their opening lines, and produce a clean transcript with each speaker's name attached. ## Quick Start Transcribe the audio file meeting-recording.mp3 with speaker names using the deepgram-transcribe skill.
