deepgram-transcription

To transcribe large video files, this solution extracts audio to reduce upload size and processing time before sending the media to the Deepgram API. It supports mp4, mov, avi, and mkv files, automatically handling the audio extraction to streamline the transcription workflow.

Deepgram transcription supports outputting both a full JSON transcript with timestamps and a plain text transcript. This allows for quick consumption of the text while retaining detailed timing data for further processing or documentation.

You can transcribe common audio formats like mp3, wav, m4a, and aac, as well as video formats including mp4, mov, avi, and mkv. The tool processes these files via the Deepgram API to produce accurate text transcripts.

You can automate transcription workflows by running the command-line script with your Deepgram API key and a specified output directory. The script handles file processing, optional smart formatting, and saves the resulting transcripts automatically.

Yes, a Deepgram API key is required to authenticate your requests and transcribe audio and video files. You must provide this key when running the transcription script to access the default nova-2 model and generate transcripts.

Yes, you can enable optional smart formatting when transcribing audio with Deepgram. This feature helps structure the plain text and JSON transcript outputs for better readability and documentation without manual editing.