Step 2: Connect the Input Preparation and Language Split

Normalize input fields and split target languages into individual items.

1. In Initialize Workflow Inputs, add assignments for the following fields:
2. Set chineseContent to {{ $json.chineseContent }}.
3. Set targetLanguages to {{ $json.targetLanguages }}.
4. Set elevenLabsApiKey to [CONFIGURE_YOUR_API_KEY].
5. Set voiceId to [YOUR_ID].
6. In Split Target Tongues, set Field To Split Out to targetLanguages and Include to allOtherFields.

Step 3: Set Up the Translation Agent and AI Model

Configure the AI translation step, language model, and structured output parser.

1. Open Multilingual Translation Agent and set Text to =Chinese content to translate: {{ $json.chineseContent }} Target language: {{ $json.targetLanguages }}.
2. Ensure Prompt Type is define and Has Output Parser is enabled.
3. In OpenAI Dialogue Model, select model gpt-4o.
4. Credential Required: Connect your openAiApi credentials in OpenAI Dialogue Model.
5. In Structured Result Parser, keep Schema Type as manual and use the provided JSON schema.
6. Note: Structured Result Parser is a sub-node of Multilingual Translation Agent; credentials (if needed) must be added to the parent’s language model node.

Step 4: Configure Audio Generation and Payload Formatting

Generate audio files in ElevenLabs and prepare standardized fields for downstream processing.

1. In ElevenLabs Audio Generator, set URL to =https://api.elevenlabs.io/v1/text-to-speech/{{ $('Initialize Workflow Inputs').item.json.voiceId }}.
2. Set Method to POST and enable Send Headers and Send Body.
3. Set JSON Body to ={ "text": {{ $json.translatedText }}, "model_id": "eleven_multilingual_v2", "voice_settings": { "stability": 0.5, "similarity_boost": 0.75, "style": 0.5, "use_speaker_boost": true } }.
4. Add header xi-api-key with value {{ $('Initialize Workflow Inputs').item.json.elevenLabsApiKey }} and Content-Type application/json.
5. In Format Audio Payload, set language to {{ $json.targetLanguage }}, translatedText to {{ $json.translatedText }}, audioFileName to {{ $json.targetLanguage }}_audio.mp3, and hasAudio to true.

Step 5: Compute Metrics, Validate Quality, and Route Outputs

Compute translation metrics, validate quality, enrich metadata, upload files, and aggregate summaries.

1. Ensure Compute Translation Metrics runs after Format Audio Payload to calculate any needed statistics.
2. Configure Validate Translation Quality with your desired IF conditions for pass/fail.
3. Validate Translation Quality outputs to both Enrich Audio Metadata and Dispatch Quality Alert in parallel.
4. In Enrich Audio Metadata, add any metadata fields required by your file storage policy.
5. In Upload Files to Drive, keep Drive set to My Drive and Folder set to root.
6. Credential Required: Connect your googleDriveOAuth2Api credentials in Upload Files to Drive.
7. After upload, Aggregate Translation Results collects items and passes them to Summarize Output Metrics, then to Assemble Final Report.
8. Credential Required: Connect your gmailOAuth2 credentials in Dispatch Quality Alert if you want failed-quality notifications.

Step 6: Test and Activate Your Workflow

Verify the end-to-end flow before turning it on in production.

1. Click Execute Workflow and send a POST request to the Inbound Webhook Start URL with chineseContent and targetLanguages.
2. Confirm that Multilingual Translation Agent returns a structured output with translatedText and targetLanguage.
3. Verify ElevenLabs Audio Generator outputs a file in the audio property and that Upload Files to Drive stores it in Google Drive.
4. Check Assemble Final Report for aggregated results and summarized metrics.
5. When successful, toggle the workflow Active to enable production runs.