AI-Powered Support Call Analysis, Logging, and Alert Routing - n8n Workflow

• Call Center Managers: Need immediate quality assurance monitoring and alerts for negative customer interactions.


• Automation Engineers: Looking for advanced n8n templates that integrate LLMs (Gemini) and specialized APIs (ElevenLabs).


• QA Specialists: Requiring automated summaries and structured data (sentiment, action items) from audio data.


• Data Analysts: Seeking centralized logging of conversational data in Google Sheets using an automated n8n node setup.

Manual review of support call recordings is time-consuming and inefficient. This high-value n8n workflow completely automates the process, providing rapid insights crucial for quality assurance and timely management intervention. The automation starts with a simple n8n trigger and moves swiftly through transcription using ElevenLabs, achieving multi-speaker diarization. The core intelligence lies in the LangChain Agent n8n node, which leverages the Google Gemini Chat Model to perform complex text analysis, extracting sentiment, speaker identification, action items, and department tags into a rigorously defined JSON structure. This structured data is immediately logged to Google Sheets and routed via a flow control n8n node (Switch), ensuring negative calls trigger manager alerts on Telegram, while positive calls send immediate kudos to the team. This powerful n8n template ensures no critical customer interaction is missed, vastly improving operational efficiency.

The entire process is initiated by an n8n trigger, specifically the Schedule Trigger, which runs every minute. First, the Search For New Call Recordings n8n node checks a specified Google Drive folder for new WAV audio files. For each file found, it is downloaded and sent via an HTTP Request n8n node to the ElevenLabs Speech-to-Text API for transcription and diarization. A subsequent Code n8n node transforms the raw API word output into a clean, conversational transcript format.

Next, the Call Analyze LangChain Agent n8n node receives this transcript. It uses the Google Gemini Chat Model and a Structured Output Parser n8n node to enforce a strict JSON schema for the analysis results. These structured results are logged into Google Sheets using the Log Recording Analysis n8n node, creating a permanent audit trail.

Following logging, the original audio file is archived by the Move Audio To Processed Folder n8n node. Finally, the Client Sentiment Switch n8n node acts as the decision point, examining the AI-generated sentiment field. If the sentiment is 'Negative', it routes the flow to send a detailed alert to managers via Telegram. If the sentiment is 'Positive', it routes to a separate Telegram n8n node to send team kudos, demonstrating effective use of flow control within the n8n workflow.

1. Import the n8n workflow: Copy the provided JSON code and paste it into your n8n instance using the 'New' -> 'Import from JSON' function.


2. Google Drive Credentials: Set up credentials for the Google Drive n8n node (used for searching, downloading, and moving files).


3. ElevenLabs Credentials: Configure the HTTP Header Auth credentials for ElevenLabs, ensuring your API key is used for the Convert Speech To Text n8n node.


4. Google Gemini Credentials: Provide the API key for the Google Gemini Chat Model n8n node.


5. Google Sheets Setup: Update the spreadsheet and sheet ID configurations in the Log Recording Analysis n8n node to point to your desired logging sheet.


6. Telegram Credentials: Configure the Telegram credentials in the Send Alert To Managers and Send Kudos to Team n8n nodes, ensuring you update the target chatId values for your specific manager and team channels.


7. Activate: Ensure the initial Schedule n8n trigger is enabled to begin the automated checking process.

Schedule Trigger (n8n trigger): The starting point of this n8n workflow, set to run every 1 minute to check for new files.
Google Drive (Search & Download n8n node): Searches for files with mimeType = 'audio/wav' in the designated call recording folder and subsequently downloads the binary data.
HTTP Request (Convert Speech To Text): Connects to the ElevenLabs API, sending the audio file as multipart form data. Key Configuration: Sets diarize: true and uses the scribev1 model to identify multiple speakers.
Code (Convert to Conversational Transcribe n8n node): Custom JavaScript code that processes the raw word-by-word transcription output from ElevenLabs, converting it into a clean, easy-to-read conversational transcript (Speaker 0: Text, Speaker 1: Text).
Google Gemini Chat Model: Provides the large language model capacity required by the Agent n8n node for advanced analysis.
Structured Output Parser (n8n node): Ensures the AI's analysis strictly adheres to a predefined JSON schema, guaranteeing consistent output for downstream logging and routing.
LangChain Agent (Call Analyze n8n node): The core analysis engine. It uses system instructions to perform complex analysis on the transcript, identifying speakers, sentiment ('Negative', 'Positive'), action items, and generating a summary.
Google Sheets (Log Recording Analysis n8n node): Appends the AI-parsed structured data (Topic, Sentiment, Actions, Fulltext) to a tracking spreadsheet.
Google Drive (Move Audio To Processed Folder n8n node): Moves the successfully processed audio file to an archival folder to prevent re-processing in the next cycle of the n8n trigger.
Switch (Client Sentiment n8n node): A crucial flow control n8n node that routes the workflow based on the value of the clientsentiment field extracted by the AI.

• Telegram (Send Alert To Managers / Send Kudos to Team n8n node): Sends personalized, sentiment-driven alerts to different internal channels, completing the notification process in this n8n workflow.