Voice Sentiment Analysis System

Wav2Vec 2.0 + BERT Pipeline

Automatically analyze customer call sentiment and classify satisfaction.

Analyze one voice call

How it works?

3-step pipeline:

  1. Audio โ†’ Text: Transcription with Wav2Vec 2.0
  2. Text โ†’ Sentiment: Analysis with multilingual BERT
  3. Classification: Customer satisfaction (Satisfied/Dissatisfied/Neutral)

Supported formats

  • WAV (recommended)
  • MP3
  • M4A

Classifications

  • ๐Ÿ˜Š Satisfied: Positive sentiment with high confidence
  • ๐Ÿ˜  Dissatisfied: Negative sentiment with high confidence
  • ๐Ÿ˜ Neutral: Neutral sentiment or low confidence

Tips

  • Clear audio quality recommended
  • Optimal duration: 10 seconds to 2 minutes
  • Avoid excessive background noise