Aihilums Sehatsanjha AI Audio Processing via Replicate - n8n Workflow

AI Developers needing reliable status checking for long-running processes.
Users requiring automated, production-ready n8n templates for audio or general AI model consumption.
Engineers looking for a robust n8n node structure for integration with Replicate or similar asynchronous APIs.
Anyone seeking to leverage the Aihilums Sehatsanjha model efficiently within a customizable automation platform.

Integrating with powerful AI models, especially those handling large data like audio, often involves asynchronous API calls. The initial request starts the process, but the results are not immediate. This robust n8n workflow provides a complete solution to this challenge. It handles the initial API prediction request, logs the details, and implements a sophisticated polling loop using various flow control n8n nodes. This specific n8n template is designed to process audio data using the aihilums/sehatsanjha model on Replicate, ensuring that the final output (success URL or detailed error) is reliably returned, making this a highly valuable n8n workflow for production use.

The execution of this powerful n8n workflow is started by the 'Manual Trigger' n8n trigger.

1. Initialization: The 'Set API Token' n8n node configures the necessary Replicate API key. The 'Set Other Parameters' n8n node then combines this token with user inputs (like audiofile, userid, and session flags) ready for the AI model.


2. Prediction Request: The 'Create Other Prediction' n8n node (an HTTP Request n8n node) sends a POST request to the Replicate API, initiating the audio processing job and obtaining a unique prediction ID.


3. Logging and Wait: The 'Log Request' n8n node logs the prediction ID for monitoring. This is followed by an initial 'Wait 5s' n8n node to give the process a head start.


4. Polling Loop: The 'Check Status' n8n node repeatedly polls the Replicate endpoint using the prediction ID to fetch the current status.


5. Flow Control: The 'Is Complete?' n8n node checks if the status is 'succeeded'. If true, it routes the successful result to the 'Success Response' n8n node. If false, it passes the data to the 'Has Failed?' n8n node.


6. Failure Handling & Retry: If the job 'Has Failed?' (status is 'failed'), the result is routed to the 'Error Response' n8n node. If neither succeeded nor failed (meaning it is still processing), the n8n workflow executes a 'Wait 10s' n8n node and loops back to 'Check Status', ensuring continuous monitoring. This core logic flow control is essential for reliable AI integrations.


7. Final Output: Both success and error paths converge at the 'Display Result' n8n node, which formats the final structured JSON output for the user.

To start using this specific n8n workflow template, follow these steps:

1. Import: Copy the entire workflow JSON and paste it into your n8n instance using the 'New' menu and selecting 'Import from JSON'.


2. Replicate API Key: Obtain your API token from your Replicate account settings (https://replicate.com).


3. Configure Credentials: Locate the 'Set API Token' n8n node.

1. Set Parameters: Navigate to the 'Set Other Parameters' n8n node to customize the inputs for the aihilums/sehatsanjha model, such as the audiofile URL or user_id.


2. Execution: Click the 'Manual Trigger' n8n trigger node and select 'Execute Workflow' to test the automation. This ready-to-use n8n template is now operational.

Manual Trigger (n8n trigger): Initiates the entire automation process when clicked. Essential for testing this n8n workflow template.
Set API Token (n8n node): Stores the private Replicate API key, dynamically accessible by subsequent HTTP Request n8n nodes for authorization.
Create Other Prediction (HTTP Request n8n node): Sends the initial, authorized request to the Replicate API, specifying the aihilums/sehatsanjha model version. It utilizes Bearer authentication and sends a structured JSON body containing all model inputs.
Wait 5s / Wait 10s (n8n node): Implements time delays necessary for asynchronous polling. The 10-second wait controls the frequency of status checks within the core logic flow control loop.
Check Status (HTTP Request n8n node): Repeatedly calls the prediction URL provided by Replicate, checking the current state of the audio processing job.
Is Complete? / Has Failed? (If n8n node): These are critical core logic flow control n8n nodes. They determine if the prediction status is 'succeeded' or 'failed', ensuring the workflow handles both successful completions and errors gracefully before retrying.