Skip to main content

Zapier Integration

The Zapier integration lets you transcribe recordings in your Zaps, without writing code. It creates jobs through the Jobs API with your API key, so the same input requirements, file types, and pricing apply.

The Zapier app​

The Speech is Cheap app for Zapier connects transcription jobs to automated workflows. Find the app by searching for Speech is Cheap in Zapier, or view its Zapier App Directory listing.

You need a Zapier account and a Speech is Cheap API key. Enter your key in the API Token field when connecting the app. How to Transcribe explains how to get a key.

Steps and actions​

The app provides two actions and one search:

  • Transcribe Audio: Creates a new transcription from an audio or video file. Pass an Input File (up to 150 MB) from an earlier step such as Google Drive or Dropbox, or an Input URL for a direct download link. Very long recordings and large files work best through Input URL. Transcribe Audio waits for transcription to complete before continuing to later steps, so a Find Transcription step is not needed directly after it.
  • Find Transcription: Reads the current status of an existing transcription by its Transcription ID (which ends in -sic) once without waiting. If the job is still processing, it returns PENDING with empty transcript fields. If the job finished, it returns COMPLETED, FAILED, or CANCELED. If the ID is unknown, it returns no match.
  • Cancel Transcription: Stops an in-progress transcription by its Transcription ID to avoid billing if the job has not completed. It requires an API key from the account that created the job. Because Transcribe Audio waits for completion, a step after Transcribe Audio in the same Zap cannot interrupt it. Use a separate workflow with a known Transcription ID instead.

For detailed input options, Advanced Options, halting rules, and editor test step behavior, see the Overview.

Outputs​

Transcription steps return:

  • Transcription ID: The job identifier ending in -sic.
  • Status: The job status (PENDING, COMPLETED, FAILED, or CANCELED). Completed Transcribe Audio actions return COMPLETED, while canceled jobs or permanent failures halt the action.
  • Transcript: Plain text formatted in paragraphs. When timestamps are enabled, each segment starts on its own line with its timestamp. Speaker parsing or Split Channels prefixes lines with labels like Speaker A:, and audio labels include non-speech sounds like [music].
  • SRT Subtitles and VTT Subtitles: Formatted subtitle text.
  • SRT File: A downloadable subtitle file object that downstream steps can save or send.
  • Detected Language: The dominant language code across all segments.
  • Duration in Seconds: Total audio duration in seconds.
  • Output Segments *: Line-item segment fields with timestamps, confidence scores, speaker IDs, audio labels, and text for workflows that process individual segments.

Guides​

GuideWhat it covers
OverviewInput File and Input URL options, Advanced Options, halting behavior, and sample transcripts in test steps.
Google Drive » Google SheetsTranscribe audio uploaded to a Google Drive folder into Google Sheets.
Google Drive » Zapier TableTranscribe audio uploaded to a Google Drive folder into a Zapier Table.
Google Drive » Google DocsSave the complete transcript as a Google Doc.
Google Drive » SRT FileSave subtitles beside the recording in Google Drive.
Google Drive » AI SummarySummarize audio uploaded to Google Drive with AI by Zapier.
  • Input Requirements covers audio length, file size, and URL rules.
  • Errors explains each Jobs API error message and how to fix it.