Automate Video Translation from Multiple Sources with Rask AI

1. Workflow Overview

This workflow automates the translation of a video accessible via URL into a target language using the Rask AI platform. It is designed for use cases such as localization, multilingual content creation, and training material preparation. It supports video sources such as YouTube, Google Drive, S3, Vimeo, or direct download links.

The workflow is logically divided into these blocks:

1.1 Input Reception and Media Upload: Receives input parameters (video link and destination language), uploads the video to Rask AI.
1.2 Media Processing Monitoring: Polls the media status until the upload and processing are complete or an error occurs.
1.3 Project Creation and Translation: Creates a translation project on Rask AI using the processed media and the target language.
1.4 Project Processing Monitoring: Polls the translation project status until the dubbing/translation is complete or an error occurs.

This design ensures robust asynchronous processing with status checks and error handling at both media upload and project translation stages.

2. Block-by-Block Analysis

2.1 Input Reception and Media Upload

Overview:
This block initializes the workflow by receiving input parameters (link and dst_lang) and uploads the video media to the Rask AI platform via its API.
Nodes Involved:
- When Executed by Another Workflow (trigger)
- Upload media (HTTP Request)
Node Details:
- When Executed by Another Workflow
  - Type: Execute Workflow Trigger (Trigger node)
  - Role: Accepts input parameters link (video URL) and dst_lang (target language code) from an external workflow or manual trigger.
  - Configuration: Disabled by default; expects inputs named link and dst_lang.
  - Inputs: External workflow or manual execution.
  - Outputs: Passes input data downstream.
  - Edge Cases: Missing or malformed inputs could cause later API calls to fail.
- Upload media
  - Type: HTTP Request
  - Role: Calls Rask AI API to upload the video media using provided link.
  - Configuration: POST to https://api.rask.ai/api/library/v1/media/link with body parameters: link (from input), kind set to "video", and name set to "n8n". OAuth2 authentication (raskOAuth2) used.
  - Inputs: Data from "When Executed by Another Workflow" node.
  - Outputs: JSON response containing uploaded media ID and status.
  - Edge Cases: Authentication failure, invalid video URL, network errors, API rate limits.

2.2 Media Processing Monitoring

Overview:
This block polls the media entity status until it reaches a terminal state: ready, failed, or still processing.
Nodes Involved:
- Get media (HTTP Request)
- Switch media status (Switch)
- Wait media processing (Wait)
- Uploading failed (Stop and Error)
Node Details:
- Get media
  - Type: HTTP Request
  - Role: Retrieves current status of uploaded media by ID from Rask AI API.
  - Configuration: GET request to https://api.rask.ai/api/library/v1/media/{{media_id}} using OAuth2 credentials.
  - Inputs: Output from "Upload media" or "Wait media processing".
  - Outputs: Current media status JSON.
  - Edge Cases: Token expiration, network issues, invalid media ID.
- Switch media status
  - Type: Switch
  - Role: Evaluates media status field to decide next action.
  - Configuration:
    - If status = "failed" → route to "Uploading failed" node.
    - If status = "ready" → route to "Create project" node.
    - If status = "processing" → route to "Wait media processing" node.
  - Inputs: Media status JSON from "Get media".
  - Outputs: One of three paths based on status.
  - Edge Cases: Unexpected status values or missing status field.
- Wait media processing
  - Type: Wait
  - Role: Introduces delay before polling media status again.
  - Configuration: Uses webhook-based wait with no explicit timeout set; waits to be triggered again.
  - Inputs: From "Switch media status" when status is "processing".
  - Outputs: Triggers "Get media" node again to re-poll status.
  - Edge Cases: Webhook trigger failures or indefinite wait if media never reaches terminal status.
- Uploading failed
  - Type: Stop and Error
  - Role: Terminates workflow with error message "Video uploading failed".
  - Inputs: From "Switch media status" if media status is "failed".
  - Outputs: None (workflow stops).
  - Edge Cases: None beyond normal error termination.

2.3 Project Creation and Translation

Overview:
Once media is ready, this block creates a translation project specifying the video ID and target language.
Nodes Involved:
- Create project (HTTP Request)
- Get project (HTTP Request)
Node Details:
- Create project
  - Type: HTTP Request
  - Role: Initiates a translation project on Rask AI with the media ID and destination language.
  - Configuration: POST to https://api.rask.ai/v2/projects with JSON body containing:
    - video_id from "Get media" node's JSON id
    - dst_lang from input dst_lang parameter
    - name set to "n8n"
      OAuth2 authentication used.
  - Inputs: From "Switch media status" node when media is ready.
  - Outputs: Project creation response JSON including project ID.
  - Edge Cases: Authentication and API errors, invalid language codes.
- Get project
  - Type: HTTP Request
  - Role: Retrieves current status of the translation project using project ID.
  - Configuration: GET request to https://api.rask.ai/v2/projects/{{project_id}} with OAuth2 credentials.
  - Inputs: Output from "Create project" node or from "Wait project processing".
  - Outputs: Project status JSON.
  - Edge Cases: Token expiration, invalid project ID, network issues.

2.4 Project Processing Monitoring

Overview:
This block polls the translation project status to monitor progress until dubbing is complete or an error occurs.
Nodes Involved:
- Switch project status (Switch)
- Wait project processing (Wait)
- Processing failed (Stop and Error)
Node Details:
- Switch project status
  - Type: Switch
  - Role: Evaluates the project status field to determine the next step.
  - Configuration:
    - If status ends with "failed" → route to "Processing failed".
    - If status equals "merging_done" → translation is complete → proceed normally (no further nodes connected here, i.e., success end).
    - Otherwise → route to "Wait project processing" to delay before next poll.
  - Inputs: Project status JSON from "Get project".
  - Outputs: One of three paths based on status.
  - Edge Cases: Unexpected status values or delays.
- Wait project processing
  - Type: Wait
  - Role: Delays before re-polling the project status.
  - Configuration: Webhook-based wait node, no explicit timeout configured.
  - Inputs: From "Switch project status" when project still processing.
  - Outputs: Triggers "Get project" node for next poll cycle.
  - Edge Cases: Possible indefinite wait if project never reaches terminal status.
- Processing failed
  - Type: Stop and Error
  - Role: Terminates workflow execution with error message "Project processing failed".
  - Inputs: From "Switch project status" if project status indicates failure.
  - Outputs: None (workflow stops).
  - Edge Cases: None beyond workflow error termination.

3. Summary Table

Node Name	Node Type	Functional Role	Input Node(s)	Output Node(s)	Sticky Note
When Executed by Another Workflow	Execute Workflow Trigger	Receives input parameters to start the workflow	(external trigger)	Upload media
Upload media	HTTP Request	Uploads video media to Rask AI via API	When Executed by Another Workflow	Get media
Get media	HTTP Request	Retrieves current media status	Upload media, Wait media processing	Switch media status	🎯 Media processing — The media entity polling is used to wait until the media is ready.
Switch media status	Switch	Routes workflow based on media status	Get media	Uploading failed, Create project, Wait media processing
Uploading failed	Stop and Error	Stops workflow on media upload failure	Switch media status	(none)
Create project	HTTP Request	Creates translation project with media ID and target language	Switch media status	Get project	🎯 Project processing — The translation project entity polling is used to wait until the project is ready. After that, the dubbing is complete and you can get all the necessary artifacts.
Get project	HTTP Request	Retrieves current translation project status	Create project, Wait project processing	Switch project status
Switch project status	Switch	Routes workflow based on project status	Get project	Processing failed, (end), Wait project processing
Processing failed	Stop and Error	Stops workflow on project processing failure	Switch project status	(none)
Wait media processing	Wait	Waits before re-polling media status	Switch media status	Get media
Wait project processing	Wait	Waits before re-polling project status	Switch project status	Get project
Sticky Note	Sticky Note	Explanation of media processing polling	(none)	(none)	🎯 Media processing — The media entity polling is used to wait until the media is ready.
Sticky Note1	Sticky Note	Explanation of project processing polling	(none)	(none)	🎯 Project processing — The translation project entity polling is used to wait until the project is ready. After that, the dubbing is complete and you can get all the necessary artifacts.
Sticky Note2	Sticky Note	Full workflow description, test inputs, and authentication instructions	(none)	(none)	Full detailed workflow description including supported sources, usage, OAuth2 requirements, and test input values.

4. Reproducing the Workflow from Scratch

Create Trigger Node: "When Executed by Another Workflow"
- Type: Execute Workflow Trigger
- Configure inputs: define two input parameters named link and dst_lang.
- Initially disable this node if manual triggering is preferred.
Create HTTP Request Node: "Upload media"
- Connect output of trigger node to this node.
- Method: POST
- URL: https://api.rask.ai/api/library/v1/media/link
- Authentication: OAuth2 with credentials named raskOAuth2.
- Body Parameters (JSON):
  - link: use expression to get value {{$json["link"]}} from trigger input
  - kind: "video" (static)
  - name: "n8n" (static)
- Enable sending body as JSON.
Create HTTP Request Node: "Get media"
- Connect output of "Upload media" to this node.
- Method: GET
- URL: use expression https://api.rask.ai/api/library/v1/media/{{$json["id"]}} to get media status by media ID from previous step.
- Authentication: OAuth2 raskOAuth2.
Create Switch Node: "Switch media status"
- Connect output of "Get media" to this node.
- Add three rules on $json.status:
  - Equals "failed" → output: "Error"
  - Equals "ready" → output: "Ready"
  - Equals "processing" → output: "Processing"
Create Stop and Error Node: "Uploading failed"
- Connect "Error" output of switch to this node.
- Set error message: "Video uploading failed".
Create HTTP Request Node: "Create project"
- Connect "Ready" output of switch to this node.
- Method: POST
- URL: https://api.rask.ai/v2/projects
- Body Parameters:
  - video_id: expression from "Get media" node {{$json["id"]}}
  - dst_lang: expression from trigger node {{$json["dst_lang"]}}
  - name: "n8n" (static)
- Authentication: OAuth2 raskOAuth2.
Create HTTP Request Node: "Get project"
- Connect output of "Create project" to this node.
- Method: GET
- URL: expression https://api.rask.ai/v2/projects/{{$json["id"]}}
- Authentication: OAuth2 raskOAuth2.
Create Switch Node: "Switch project status"
- Connect output of "Get project" to this node.
- Add three rules on $json.status:
  - Ends with "failed" → output: "Error"
  - Equals "merging_done" → output: "Ready"
  - Otherwise → output: "Processing"
Create Stop and Error Node: "Processing failed"
- Connect "Error" output of switch to this node.
- Set error message: "Project processing failed".
Create Wait Node: "Wait media processing"
- Connect "Processing" output of "Switch media status" to this node.
- Use webhook-based wait with default parameters.
- Connect output of this node back to "Get media" to poll again.
Create Wait Node: "Wait project processing"
- Connect "Processing" output of "Switch project status" to this node.
- Use webhook-based wait with default parameters.
- Connect output of this node back to "Get project" to poll again.
Connect node outputs as described in the connections above.
Set up OAuth2 Credentials in n8n:
- Name: raskOAuth2
- Provide Client ID and Client Secret from your Rask AI account settings (https://app.rask.ai/account).
- Configure OAuth2 scopes and token URLs as per Rask AI documentation (https://docs.api.rask.ai/api-reference/authentication).
Add Sticky Note nodes for documentation inside the workflow:
- Add notes explaining media processing, project processing, and overall workflow context with links to Rask API docs and test inputs.

5. General Notes & Resources

Note Content	Context or Link
OAuth2 credentials for Rask AI must include Client ID and Client Secret, accessible via https://app.rask.ai/account.	Rask AI account settings
Official Rask AI API Authentication documentation: https://docs.api.rask.ai/api-reference/authentication	API docs for OAuth2 setup
Media entity polling documentation: https://docs.api.rask.ai/api-reference/media/get_media	Used for monitoring media processing status
Translation project polling documentation: https://docs.api.rask.ai/api-reference/project/get_project	Used for monitoring project translation status
Supported video sources include YouTube, Google Drive, S3, Vimeo, or direct links. Destination languages supported are listed here: https://docs.api.rask.ai/languages/destination	Workflow usage and limitations
Test inputs for quick workflow run: `link`: https://static.rask.ai/samples/sample_1.mp4, `dst_lang`: `en-us`	Provided for quick testing
This workflow is designed to be triggered by another workflow or manually, enabling integration into larger automation pipelines.	Workflow design note

Disclaimer:
The text provided comes exclusively from an automated workflow created with n8n, an integration and automation tool. This processing strictly adheres to current content policies and contains no illegal, offensive, or protected elements. All manipulated data is legal and publicly accessible.

21 KiB Raw Blame History