Sentiment Analysis

Sentiment analysis automatically detects the emotional tone of transcribed content using AI intelligence. It classifies text as positive, negative, or neutral, and provides a confidence score indicating the strength of the sentiment.

How to Enable

"enable_sentiment_analysis": "true"

During Transcription

Parameters

• enable_sentiment_analysis (required): Set to "true"

Request

import requests

url = "https://tb2.shunyalabs.ai/v1/transcriptions"
headers = {"X-API-Key": "your-api-key"}

with open("customer_feedback.wav", "rb") as f:
    files = {"file": f}
    data = {
        "enable_sentiment_analysis": "true"
    }

response = requests.post(
    url,
    headers=headers,
    files=files,
    data=data
)

print(response.json())

Example Output

{
  "success": true,
  "text": "I absolutely love this product! The customer service team was incredibly helpful and resolved my issue within minutes. This is exactly what I was looking for.",
  "segments": [...],
  "sentiment": {
    "sentiment": "positive",
    "score": 0.92
  }
}

Standalone Sentiment Analysis

Parameters

• text (required): Input text to analyze

Request

import requests

url = "https://tb.shunyalabs.ai/v1/sentiment"
headers = {"X-API-Key": "your-api-key"}

data = {
    "text": "I'm extremely frustrated with this service. I've been waiting for three weeks and still haven't received any response."
}

response = requests.post(url, headers=headers, data=data)
print(response.json())

Example Output

{
  "sentiment": "negative",
  "score": 0.88
}

Understanding Sentiment Scores

score (0.0 – 1.0) indicates confidence in the sentiment classification.

• Scores > 0.8: High confidence
• Scores 0.6 – 0.8: Moderate confidence
• Scores < 0.6: Lower confidence (mixed or ambiguous sentiment)

Best Practices

• Provide sufficient context (avoid very short inputs)
• Mixed emotions often result in neutral sentiment
• Best used for overall tone, not word-by-word analysis
• Combine with intent detection for deeper insights

Use Cases

• Customer satisfaction monitoring in support calls
• Brand and product review analysis
• Employee engagement and feedback analysis
• Market research and focus groups
• Quality assurance and risk detection
• Tracking reactions to product launches