backend/models/transcript_segment.py

"""
TranscriptSegment model for storing real-time interview transcription data
"""
from sqlalchemy import Column, Integer, String, Float, Text, DateTime, ForeignKey
from sqlalchemy.sql import func
from sqlalchemy.orm import relationship
from .database import Base


class TranscriptSegment(Base):
    __tablename__ = 'transcript_segments'
    
    id = Column(Integer, primary_key=True, index=True)
    interview_id = Column(Integer, ForeignKey('interviews.id'), nullable=False, index=True)
    timestamp = Column(Float, nullable=False)  # Seconds from interview start
    speaker = Column(String(50), nullable=False)  # 'candidate' or 'interviewer'
    text = Column(Text, nullable=False)
    confidence = Column(Float)  # Whisper confidence score (0.0 to 1.0)
    created_at = Column(DateTime(timezone=True), server_default=func.now())
    
    # Relationships
    interview = relationship('Interview', back_populates='transcript_segments')
    
    def __repr__(self):
        return f"<TranscriptSegment(id={self.id}, interview_id={self.interview_id}, speaker='{self.speaker}', timestamp={self.timestamp})>"
    
    def to_dict(self):
        return {
            'id': self.id,
            'interview_id': self.interview_id,
            'timestamp': self.timestamp,
            'speaker': self.speaker,
            'text': self.text,
            'confidence': self.confidence,
            'created_at': self.created_at.isoformat() if self.created_at else None
        }