ASSEMBLYAI n8n INTEGRATION: AUTOMATE ASSEMBLYAI WITH N8N

This action searches through a transcript for specific words or phrases, returning timestamps and context—essential for building searchable audio archives or finding specific moments in recordings.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (set to "Transcript" - required), Operation (set to "Word Search" - required), Transcript ID (the transcript to search within - required), Input Type (accepts comma-separated lists), Words to Search (comma-separated list of terms - accepts dynamic expressions - required).

Use cases: Finding mentions of competitor names in sales calls, locating specific topics in long recordings, building "jump to" features for audio players, compliance monitoring for regulated terms.

For privacy and compliance, this action permanently removes data associated with a specific LeMUR request from AssemblyAI's systems.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (set to "LeMUR" - required), Operation (set to "Purge Data" - required), LeMUR Request ID (the request whose data should be deleted - optional).

Use cases: GDPR compliance workflows, removing sensitive analysis results, data retention policy enforcement, cleaning up after temporary processing.

The most flexible LeMUR action—send any custom prompt to analyze your transcripts however you need, from sentiment analysis to structured data extraction.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (LeMUR resource - required), Operation (custom task execution - required), Transcript IDs (one or more transcripts to process - optional), Final Model (your chosen AI model like Claude 4 Sonnet - required), Prompt (your custom instructions for analyzing the transcript - required), Max Output Size (character limit for response, defaults to 2000 - optional).

Use cases: Custom sentiment analysis, extracting structured JSON data from conversations, generating social media posts from audio content, creating custom report formats, multi-language translation of key points.

Powered by AssemblyAI's LeMUR (Large Language Model Understanding and Reasoning), this action creates intelligent summaries of your transcripts—far more sophisticated than simple truncation.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (typically LeMUR-based - required), Operation (set to generate summaries - required), Transcript IDs (one or more transcript IDs to summarize - required), Final Model (select the AI model for processing like Claude 4 Sonnet - required), Temperature (controls output randomness 0-1, lower values produce more consistent summaries - optional), Max Output Size (maximum token length for generated summary - optional).

Use cases: Creating podcast episode descriptions, generating meeting recap emails, producing executive summaries of long calls, extracting key points for newsletters.

This LeMUR-powered action lets you ask natural language questions about your transcript content and receive accurate, context-aware answers.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (LeMUR resource - required), Operation (set to "Question & Answer" - required), Transcript IDs (the transcripts to analyze - required), Final Model (the AI model to use for answering - required), Questions (the specific questions you want answered about the content), Temperature (controls answer creativity vs. precision - optional).

Use cases: Extracting action items from meeting recordings, answering specific queries about sales calls ("What objections did the prospect raise?"), analyzing interview responses, pulling specific data points from audio content.

When running LeMUR operations asynchronously, this action retrieves the response from a previous request using its unique ID.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (set to "LeMUR" - required), Operation (set to "Get Response" - required), LeMUR Request ID (the unique identifier from your original LeMUR request - required).

Use cases: Polling for completed LeMUR analyses, building async workflow patterns, handling long-running AI processing jobs.

When you need to remove transcription data—whether for privacy compliance, storage management, or cleanup—this action permanently deletes a specified transcript from AssemblyAI.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (set to "Transcript" - required), Operation (set to "Delete" - required), Transcript ID (the unique identifier of the transcript to remove - required).

Use cases: Implementing data retention policies, removing test transcriptions, GDPR compliance workflows, cleaning up after temporary processing jobs.

When transcripts contain sensitive information (PII, profanity, custom terms), this action retrieves an audio file where those elements have been replaced with neutral tones—perfect for privacy-conscious applications.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (set to "Transcript" - required), Operation (set to "Get Redacted Audio" - required), Transcript ID (the transcript with PII redaction enabled - required).

Use cases: Creating shareable versions of customer calls, removing sensitive data from training recordings, preparing compliant audio for public distribution, anonymizing interview recordings.

This action retrieves a list of all your transcriptions, making it easy to audit processing history, find specific transcripts, or build dashboards showing transcription activity.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (set to "Transcript" - required), Operation (set to "List" - required), Limit (numeric value controlling how many transcriptions to return, defaults to 50 - required), Additional Fields (optional section for adding filtering parameters).

Use cases: Building transcription monitoring dashboards, auditing API usage, finding transcripts from a specific date range, creating reports on transcription volume.

Once you've created a transcription job, this action retrieves the complete results. Since transcription is asynchronous (it takes time to process), you'll typically poll this endpoint or use it after a delay.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (set to "Transcript" - required), Operation (set to "Get" - required), Transcript ID (the unique identifier returned when you created the transcription, accepts both fixed values and dynamic expressions from previous nodes - required).

Use cases: Checking transcription status in a polling loop, retrieving completed transcripts after a Wait node delay, fetching transcript data for downstream processing.

This action breaks down a transcript into individual sentences with precise timing information—essential for creating searchable content or detailed analysis.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (set to "Transcript" - required), Operation (set to "Get Sentences" - required), Transcript ID (the transcript you want to segment into sentences - required).

Use cases: Creating timestamped show notes for podcasts, building searchable video indices, extracting specific quotes with timing references, generating chapter markers.

This action initiates the core transcription process—taking an audio file URL and returning a complete text transcript with timestamps, speaker labels, and additional analysis.

Key parameters: Credential to connect with (your AssemblyAI account selection - required), Resource (set to "Transcript" - required), Operation (set to "Create" - required), Audio URL (the publicly accessible URL of your audio file, can be an AssemblyAI upload URL from the previous action or any public audio link - required), Result (output field that captures the transcription job response, including the transcript ID needed for subsequent operations).

Use cases: Transcribing podcast episodes from RSS feed URLs, processing recorded Zoom meetings, converting voice memos to searchable text, generating transcripts for YouTube video audio tracks.

This action generates properly formatted subtitle files from your transcripts—ready for use with video players, YouTube uploads, or accessibility compliance.

Key parameters: Credential to connect with (your AssemblyAI account - required), Resource (set to "Transcript" - required), Operation (set to "Get Subtitles" - required), Transcript ID (the source transcript for subtitle generation, accepts text input and expressions - required), Result (displays the outcome after execution), Characters Per Caption (numeric value setting maximum characters per caption line for readability - required).

Use cases: Automating YouTube video captioning, creating SRT files for video editing software, generating accessibility captions for course content, producing subtitles for social media videos.

Similar to sentences but at a higher level—this action retrieves transcript content organized into logical paragraph segments, providing natural groupings of related content.

Key parameters: Credential to connect with (your AssemblyAI credentials - required), Resource (set to "Transcript" - required), Operation (set to "Get Paragraphs" - required), Transcript ID (the transcript to retrieve paragraphs from - required).

Use cases: Generating formatted blog posts from audio content, creating readable meeting summaries, structuring content for documentation, preparing text for translation services.

The Upload a file action serves as the starting point for many AssemblyAI workflows—it lets you send audio files directly to AssemblyAI's servers before transcription. This is particularly useful when your audio source isn't publicly accessible via URL, similar to how you might handle files in Google Drive integrations.

Key parameters: Credential to connect with (select your configured AssemblyAI account - required), Resource (set to "File" to indicate you're working with file uploads - required), Operation (automatically set to "Upload" - required), File Path or Binary Data (accepts either a direct file path string or binary data from previous nodes, perfect for processing files from Google Drive, Dropbox, or HTTP downloads - required).

Use cases: Uploading private recordings from cloud storage, processing audio files received via email attachments, handling locally stored audio from FTP servers.