Skip to main content
This example demonstrates how to perform sentiment analysis on audio conversations using Agno agents with multimodal capabilities.
audio_sentiment_analysis.py
import requests
from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.media import Audio
from agno.models.google import Gemini

db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
agent = Agent(
    model=Gemini(id="gemini-2.0-flash-exp"),
    add_history_to_context=True,
    markdown=True,
    db=SqliteDb(
        session_table="audio_sentiment_analysis_sessions",
        db_file="tmp/audio_sentiment_analysis.db",
    ),
)

url = "https://agno-public.s3.amazonaws.com/demo_data/sample_conversation.wav"

response = requests.get(url)
audio_content = response.content

# Give a sentiment analysis of this audio conversation. Use speaker A, speaker B to identify speakers.
agent.print_response(
    "Give a sentiment analysis of this audio conversation. Use speaker A, speaker B to identify speakers.",
    audio=[Audio(content=audio_content)],
    stream=True,
)

agent.print_response(
    "What else can you tell me about this audio conversation?",
    stream=True,
)

​
Key Features

  • Audio Processing: Downloads and processes audio files from remote URLs
  • Sentiment Analysis: Analyzes emotional tone and sentiment in conversations
  • Speaker Identification: Distinguishes between different speakers in the conversation
  • Persistent Sessions: Maintains conversation history using SQLite database
  • Streaming Response: Real-time response generation for better user experience

​
Use Cases

  • Customer service call analysis
  • Meeting sentiment tracking
  • Interview evaluation
  • Call center quality monitoring
⌘I