karaoke-gen 0.57.0__py3-none-any.whl → 0.71.27__py3-none-any.whl

lyrics_transcriber/frontend/src/components/AudioPlayer.tsx

@@ -0,0 +1,180 @@
+import { Box, IconButton, Slider, Typography } from '@mui/material'
+import PlayArrowIcon from '@mui/icons-material/PlayArrow'
+import PauseIcon from '@mui/icons-material/Pause'
+import { useEffect, useRef, useState, useCallback } from 'react'
+import { ApiClient } from '../api'
+
+interface AudioPlayerProps {
+    apiClient: ApiClient | null,
+    onTimeUpdate?: (time: number) => void,
+    audioHash: string
+}
+
+export default function AudioPlayer({ apiClient, onTimeUpdate, audioHash }: AudioPlayerProps) {
+    const [isPlaying, setIsPlaying] = useState(false)
+    const [currentTime, setCurrentTime] = useState(0)
+    const [duration, setDuration] = useState(0)
+    const audioRef = useRef<HTMLAudioElement | null>(null)
+
+    useEffect(() => {
+        if (!apiClient) return
+
+        const audio = new Audio(apiClient.getAudioUrl(audioHash))
+        audioRef.current = audio
+
+        // Add requestAnimationFrame for smoother updates
+        let animationFrameId: number
+
+        const updateTime = () => {
+            const time = audio.currentTime
+            setCurrentTime(time)
+            onTimeUpdate?.(time)
+            animationFrameId = requestAnimationFrame(updateTime)
+        }
+
+        audio.addEventListener('play', () => {
+            setIsPlaying(true)
+            window.isAudioPlaying = true
+            updateTime()
+        })
+
+        audio.addEventListener('pause', () => {
+            setIsPlaying(false)
+            window.isAudioPlaying = false
+            cancelAnimationFrame(animationFrameId)
+        })
+
+        audio.addEventListener('ended', () => {
+            cancelAnimationFrame(animationFrameId)
+            setIsPlaying(false)
+            window.isAudioPlaying = false
+            setCurrentTime(0)
+        })
+
+        audio.addEventListener('loadedmetadata', () => {
+            setDuration(audio.duration)
+        })
+
+        return () => {
+            cancelAnimationFrame(animationFrameId)
+            audio.pause()
+            audio.src = ''
+            audioRef.current = null
+            window.isAudioPlaying = false
+        }
+    }, [apiClient, onTimeUpdate, audioHash])
+
+    const handlePlayPause = () => {
+        if (!audioRef.current) return
+
+        if (isPlaying) {
+            audioRef.current.pause()
+        } else {
+            audioRef.current.play()
+        }
+        setIsPlaying(!isPlaying)
+    }
+
+    const handleSeek = (_: Event, newValue: number | number[]) => {
+        if (!audioRef.current) return
+        const time = newValue as number
+        audioRef.current.currentTime = time
+        setCurrentTime(time)
+    }
+
+    const formatTime = (seconds: number) => {
+        const mins = Math.floor(seconds / 60)
+        const secs = Math.floor(seconds % 60)
+        return `${mins}:${secs.toString().padStart(2, '0')}`
+    }
+
+    // Add this method to expose seeking functionality
+    const seekAndPlay = (time: number) => {
+        if (!audioRef.current) return
+
+        audioRef.current.currentTime = time
+        setCurrentTime(time)
+        audioRef.current.play()
+        setIsPlaying(true)
+    }
+
+    const togglePlayback = useCallback(() => {
+        if (!audioRef.current) return
+
+        if (isPlaying) {
+            audioRef.current.pause()
+        } else {
+            audioRef.current.play()
+        }
+        setIsPlaying(!isPlaying)
+    }, [isPlaying])
+
+    // Expose methods and duration globally
+    useEffect(() => {
+        if (!apiClient) return
+
+        // eslint-disable-next-line @typescript-eslint/no-explicit-any
+        const win = window as any
+        win.seekAndPlayAudio = seekAndPlay
+        win.toggleAudioPlayback = togglePlayback
+        win.getAudioDuration = () => duration
+
+        return () => {
+            delete win.seekAndPlayAudio
+            delete win.toggleAudioPlayback
+            delete win.getAudioDuration
+        }
+    }, [apiClient, togglePlayback, duration])
+
+    if (!apiClient) return null
+
+    return (
+        <Box sx={{
+            display: 'flex',
+            alignItems: 'center',
+            gap: 0.5,
+            backgroundColor: 'background.paper',
+            borderRadius: 1,
+            height: '32px',
+        }}>
+            <Typography variant="body2" color="text.secondary" sx={{ mr: 0.5, fontSize: '0.75rem' }}>
+                Playback:
+            </Typography>
+
+            <IconButton
+                onClick={handlePlayPause}
+                size="small"
+                sx={{ p: 0.5 }}
+            >
+                {isPlaying ? <PauseIcon fontSize="small" /> : <PlayArrowIcon fontSize="small" />}
+            </IconButton>
+
+            <Typography variant="body2" sx={{ minWidth: 32, fontSize: '0.75rem' }}>
+                {formatTime(currentTime)}
+            </Typography>
+
+            <Slider
+                value={currentTime}
+                min={0}
+                max={duration}
+                onChange={handleSeek}
+                size="small"
+                sx={{
+                    width: 150,
+                    mx: 0.5,
+                    '& .MuiSlider-thumb': {
+                        width: 10,
+                        height: 10,
+                    },
+                    '& .MuiSlider-rail, & .MuiSlider-track': {
+                        height: 3
+                    }
+                }}
+            />
+
+            <Typography variant="body2" sx={{ minWidth: 32, fontSize: '0.75rem' }}>
+                {formatTime(duration)}
+            </Typography>
+        </Box>
+    )
+} 

lyrics_transcriber/frontend/src/components/CorrectedWordWithActions.tsx

@@ -0,0 +1,167 @@
+import { Box, IconButton, Tooltip, useMediaQuery, useTheme } from '@mui/material'
+import UndoIcon from '@mui/icons-material/Undo'
+import EditIcon from '@mui/icons-material/Edit'
+import CheckCircleOutlineIcon from '@mui/icons-material/CheckCircleOutline'
+import { COLORS } from './shared/constants'
+import { styled } from '@mui/material/styles'
+
+interface CorrectionInfo {
+    originalWord: string
+    handler: string
+    confidence: number
+    source: string
+    reason?: string
+}
+
+interface CorrectedWordWithActionsProps {
+    word: string
+    originalWord: string
+    correction: CorrectionInfo
+    onRevert: () => void
+    onEdit: () => void
+    onAccept: () => void
+    onClick?: () => void
+    backgroundColor?: string
+    shouldFlash?: boolean
+}
+
+const WordContainer = styled(Box, {
+    shouldForwardProp: (prop) => !['shouldFlash'].includes(prop as string)
+})<{ shouldFlash?: boolean }>(({ shouldFlash }) => ({
+    display: 'inline-flex',
+    alignItems: 'center',
+    gap: '2px',
+    padding: '1px 3px',
+    borderRadius: '2px',
+    cursor: 'pointer',
+    position: 'relative',
+    backgroundColor: COLORS.corrected,
+    animation: shouldFlash ? 'flash 1s ease-in-out infinite' : 'none',
+    '@keyframes flash': {
+        '0%, 100%': { opacity: 1 },
+        '50%': { opacity: 0.5 }
+    },
+    '&:hover': {
+        backgroundColor: '#c8e6c9'
+    }
+}))
+
+const OriginalWordLabel = styled(Box)({
+    position: 'absolute',
+    top: '-14px',
+    left: '0',
+    fontSize: '0.6rem',
+    color: '#666',
+    textDecoration: 'line-through',
+    opacity: 0.7,
+    whiteSpace: 'nowrap',
+    pointerEvents: 'none'
+})
+
+const ActionsContainer = styled(Box)({
+    display: 'inline-flex',
+    alignItems: 'center',
+    gap: '1px',
+    marginLeft: '2px'
+})
+
+const ActionButton = styled(IconButton)(({ theme }) => ({
+    padding: '2px',
+    minWidth: '20px',
+    minHeight: '20px',
+    width: '20px',
+    height: '20px',
+    backgroundColor: 'rgba(255, 255, 255, 0.9)',
+    border: '1px solid rgba(0, 0, 0, 0.1)',
+    '&:hover': {
+        backgroundColor: 'rgba(255, 255, 255, 1)',
+        transform: 'scale(1.1)'
+    },
+    '& .MuiSvgIcon-root': {
+        fontSize: '0.875rem'
+    },
+    // Ensure minimum touch target on mobile
+    [theme.breakpoints.down('sm')]: {
+        minWidth: '28px',
+        minHeight: '28px',
+        width: '28px',
+        height: '28px',
+        padding: '4px'
+    }
+}))
+
+export default function CorrectedWordWithActions({
+    word,
+    originalWord,
+    onRevert,
+    onEdit,
+    onAccept,
+    onClick,
+    backgroundColor,
+    shouldFlash
+}: CorrectedWordWithActionsProps) {
+    const theme = useTheme()
+    const isMobile = useMediaQuery(theme.breakpoints.down('sm'))
+
+    const handleAction = (e: React.MouseEvent, action: () => void) => {
+        e.stopPropagation()
+        action()
+    }
+
+    return (
+        <WordContainer
+            shouldFlash={shouldFlash}
+            sx={{ backgroundColor: backgroundColor || COLORS.corrected }}
+            onClick={onClick}
+        >
+            <OriginalWordLabel>{originalWord}</OriginalWordLabel>
+            
+            <Box
+                component="span"
+                sx={{
+                    fontSize: '0.85rem',
+                    lineHeight: 1.2,
+                    fontWeight: 600
+                }}
+            >
+                {word}
+            </Box>
+
+            <ActionsContainer>
+                <Tooltip title="Revert to original" placement="top" arrow>
+                    <ActionButton
+                        size="small"
+                        onClick={(e) => handleAction(e, onRevert)}
+                        aria-label="revert correction"
+                    >
+                        <UndoIcon />
+                    </ActionButton>
+                </Tooltip>
+
+                <Tooltip title="Edit correction" placement="top" arrow>
+                    <ActionButton
+                        size="small"
+                        onClick={(e) => handleAction(e, onEdit)}
+                        aria-label="edit correction"
+                    >
+                        <EditIcon />
+                    </ActionButton>
+                </Tooltip>
+
+                {!isMobile && (
+                    <Tooltip title="Accept correction" placement="top" arrow>
+                        <ActionButton
+                            size="small"
+                            onClick={(e) => handleAction(e, onAccept)}
+                            aria-label="accept correction"
+                            sx={{ color: 'success.main' }}
+                        >
+                            <CheckCircleOutlineIcon />
+                        </ActionButton>
+                    </Tooltip>
+                )}
+            </ActionsContainer>
+        </WordContainer>
+    )
+}
+

lyrics_transcriber/frontend/src/components/CorrectionAnnotationModal.tsx

@@ -0,0 +1,359 @@
+import { useState, useEffect } from 'react'
+import {
+    Dialog,
+    DialogTitle,
+    DialogContent,
+    DialogActions,
+    Button,
+    FormControl,
+    InputLabel,
+    Select,
+    MenuItem,
+    TextField,
+    Slider,
+    Typography,
+    Box,
+    Paper,
+    Chip,
+    Alert,
+    SelectChangeEvent
+} from '@mui/material'
+import { CorrectionAnnotation, CorrectionAnnotationType, CorrectionAction } from '../types'
+
+interface CorrectionAnnotationModalProps {
+    open: boolean
+    onClose: () => void
+    onSave: (annotation: Omit<CorrectionAnnotation, 'annotation_id' | 'timestamp'>) => void
+    onSkip: () => void
+    // Context about the correction being made
+    originalText: string
+    correctedText: string
+    wordIdsAffected: string[]
+    // Optional: What the AI suggested
+    agenticProposal?: {
+        action: string
+        replacement_text?: string
+        confidence: number
+        reason: string
+        gap_category?: string
+    }
+    // Reference lyrics that were consulted
+    referenceSources: string[]
+    // Song metadata
+    audioHash: string
+    artist: string
+    title: string
+    sessionId: string
+    gapId?: string
+}
+
+const ANNOTATION_TYPES: { value: CorrectionAnnotationType; label: string; description: string }[] = [
+    { 
+        value: 'SOUND_ALIKE' as CorrectionAnnotationType, 
+        label: 'Sound-Alike Error', 
+        description: 'Homophones or similar-sounding words (e.g., "out" vs "now")' 
+    },
+    { 
+        value: 'BACKGROUND_VOCALS' as CorrectionAnnotationType, 
+        label: 'Background Vocals', 
+        description: 'Backing vocals that should be removed from karaoke' 
+    },
+    { 
+        value: 'EXTRA_WORDS' as CorrectionAnnotationType, 
+        label: 'Extra Filler Words', 
+        description: 'Transcription added words like "And", "But", "Well"' 
+    },
+    { 
+        value: 'PUNCTUATION_ONLY' as CorrectionAnnotationType, 
+        label: 'Punctuation/Style Only', 
+        description: 'Only punctuation or capitalization differences' 
+    },
+    { 
+        value: 'REPEATED_SECTION' as CorrectionAnnotationType, 
+        label: 'Repeated Section', 
+        description: 'Chorus or verse repetition not in condensed references' 
+    },
+    { 
+        value: 'COMPLEX_MULTI_ERROR' as CorrectionAnnotationType, 
+        label: 'Complex Multi-Error', 
+        description: 'Multiple different error types in one section' 
+    },
+    { 
+        value: 'AMBIGUOUS' as CorrectionAnnotationType, 
+        label: 'Ambiguous', 
+        description: 'Unclear without listening to audio' 
+    },
+    { 
+        value: 'NO_ERROR' as CorrectionAnnotationType, 
+        label: 'No Error', 
+        description: 'Transcription matches at least one reference source' 
+    },
+    { 
+        value: 'MANUAL_EDIT' as CorrectionAnnotationType, 
+        label: 'Manual Edit', 
+        description: 'Human-initiated correction not from detected gap' 
+    }
+]
+
+const CONFIDENCE_LABELS: { [key: number]: string } = {
+    1: '1 - Very Uncertain',
+    2: '2 - Somewhat Uncertain',
+    3: '3 - Neutral',
+    4: '4 - Fairly Confident',
+    5: '5 - Very Confident'
+}
+
+export default function CorrectionAnnotationModal({
+    open,
+    onClose,
+    onSave,
+    onSkip,
+    originalText,
+    correctedText,
+    wordIdsAffected,
+    agenticProposal,
+    referenceSources,
+    audioHash,
+    artist,
+    title,
+    sessionId,
+    gapId
+}: CorrectionAnnotationModalProps) {
+    const [annotationType, setAnnotationType] = useState<CorrectionAnnotationType>('MANUAL_EDIT' as CorrectionAnnotationType)
+    const [actionTaken, setActionTaken] = useState<CorrectionAction>('REPLACE' as CorrectionAction)
+    const [confidence, setConfidence] = useState<number>(3)
+    const [reasoning, setReasoning] = useState<string>('')
+    const [agenticAgreed, setAgenticAgreed] = useState<boolean>(false)
+    const [error, setError] = useState<string>('')
+
+    // Pre-fill if we have an agentic proposal
+    useEffect(() => {
+        if (agenticProposal && agenticProposal.gap_category) {
+            setAnnotationType(agenticProposal.gap_category as CorrectionAnnotationType)
+            // Check if human correction matches AI suggestion
+            const agreed = agenticProposal.replacement_text 
+                ? correctedText.toLowerCase().includes(agenticProposal.replacement_text.toLowerCase())
+                : false
+            setAgenticAgreed(agreed)
+        }
+    }, [agenticProposal, correctedText])
+
+    // Determine action type based on correction
+    useEffect(() => {
+        if (originalText === correctedText) {
+            setActionTaken('NO_ACTION' as CorrectionAction)
+        } else if (correctedText === '') {
+            setActionTaken('DELETE' as CorrectionAction)
+        } else if (originalText === '') {
+            setActionTaken('INSERT' as CorrectionAction)
+        } else {
+            setActionTaken('REPLACE' as CorrectionAction)
+        }
+    }, [originalText, correctedText])
+
+    const handleSave = () => {
+        // Validate
+        if (reasoning.length < 10) {
+            setError('Reasoning must be at least 10 characters')
+            return
+        }
+
+        const annotation: Omit<CorrectionAnnotation, 'annotation_id' | 'timestamp'> = {
+            audio_hash: audioHash,
+            gap_id: gapId || null,
+            annotation_type: annotationType,
+            action_taken: actionTaken,
+            original_text: originalText,
+            corrected_text: correctedText,
+            confidence,
+            reasoning,
+            word_ids_affected: wordIdsAffected,
+            agentic_proposal: agenticProposal || null,
+            agentic_category: agenticProposal?.gap_category || null,
+            agentic_agreed: agenticAgreed,
+            reference_sources_consulted: referenceSources,
+            artist,
+            title,
+            session_id: sessionId
+        }
+
+        onSave(annotation)
+        handleClose()
+    }
+
+    const handleClose = () => {
+        // Reset form
+        setAnnotationType('MANUAL_EDIT' as CorrectionAnnotationType)
+        setConfidence(3)
+        setReasoning('')
+        setError('')
+        setAgenticAgreed(false)
+        onClose()
+    }
+
+    const handleSkipAndClose = () => {
+        handleClose()
+        onSkip()
+    }
+
+    return (
+        <Dialog 
+            open={open} 
+            onClose={handleClose}
+            maxWidth="md"
+            fullWidth
+        >
+            <DialogTitle>
+                <Typography variant="h6">Annotate Your Correction</Typography>
+                <Typography variant="caption" color="text.secondary">
+                    Help improve the AI by explaining your correction
+                </Typography>
+            </DialogTitle>
+
+            <DialogContent dividers>
+                {/* Show what changed */}
+                <Paper elevation={0} sx={{ p: 2, mb: 3, bgcolor: 'grey.50' }}>
+                    <Typography variant="subtitle2" gutterBottom>
+                        Your Correction:
+                    </Typography>
+                    <Box sx={{ display: 'flex', gap: 2, alignItems: 'center' }}>
+                        <Box sx={{ flex: 1 }}>
+                            <Typography variant="caption" color="error">Original:</Typography>
+                            <Typography 
+                                sx={{ 
+                                    textDecoration: 'line-through', 
+                                    color: 'error.main',
+                                    fontFamily: 'monospace'
+                                }}
+                            >
+                                {originalText || '(empty)'}
+                            </Typography>
+                        </Box>
+                        <Typography variant="h6">→</Typography>
+                        <Box sx={{ flex: 1 }}>
+                            <Typography variant="caption" color="success.main">Corrected:</Typography>
+                            <Typography 
+                                sx={{ 
+                                    color: 'success.main',
+                                    fontWeight: 'bold',
+                                    fontFamily: 'monospace'
+                                }}
+                            >
+                                {correctedText || '(empty)'}
+                            </Typography>
+                        </Box>
+                    </Box>
+                </Paper>
+
+                {/* Show AI suggestion if available */}
+                {agenticProposal && (
+                    <Alert 
+                        severity={agenticAgreed ? "success" : "info"} 
+                        sx={{ mb: 3 }}
+                    >
+                        <Typography variant="subtitle2" gutterBottom>
+                            AI Suggestion:
+                        </Typography>
+                        <Typography variant="body2">
+                            Category: <strong>{agenticProposal.gap_category}</strong>
+                        </Typography>
+                        <Typography variant="body2">
+                            Action: <strong>{agenticProposal.action}</strong>
+                            {agenticProposal.replacement_text && ` → "${agenticProposal.replacement_text}"`}
+                        </Typography>
+                        <Typography variant="body2">
+                            Reason: {agenticProposal.reason}
+                        </Typography>
+                        <Typography variant="body2" sx={{ mt: 1 }}>
+                            {agenticAgreed 
+                                ? '✓ Your correction matches the AI suggestion' 
+                                : '✗ Your correction differs from the AI suggestion'}
+                        </Typography>
+                    </Alert>
+                )}
+
+                {/* Reference sources */}
+                {referenceSources.length > 0 && (
+                    <Box sx={{ mb: 3 }}>
+                        <Typography variant="subtitle2" gutterBottom>
+                            Reference Sources Consulted:
+                        </Typography>
+                        <Box sx={{ display: 'flex', gap: 1, flexWrap: 'wrap' }}>
+                            {referenceSources.map(source => (
+                                <Chip key={source} label={source} size="small" />
+                            ))}
+                        </Box>
+                    </Box>
+                )}
+
+                {/* Annotation type */}
+                <FormControl fullWidth sx={{ mb: 3 }}>
+                    <InputLabel>Correction Type *</InputLabel>
+                    <Select
+                        value={annotationType}
+                        label="Correction Type *"
+                        onChange={(e: SelectChangeEvent) => setAnnotationType(e.target.value as CorrectionAnnotationType)}
+                    >
+                        {ANNOTATION_TYPES.map(type => (
+                            <MenuItem key={type.value} value={type.value}>
+                                <Box>
+                                    <Typography variant="body1">{type.label}</Typography>
+                                    <Typography variant="caption" color="text.secondary">
+                                        {type.description}
+                                    </Typography>
+                                </Box>
+                            </MenuItem>
+                        ))}
+                    </Select>
+                </FormControl>
+
+                {/* Confidence slider */}
+                <Box sx={{ mb: 3 }}>
+                    <Typography variant="subtitle2" gutterBottom>
+                        Confidence in Your Correction *
+                    </Typography>
+                    <Slider
+                        value={confidence}
+                        onChange={(_, newValue) => setConfidence(newValue as number)}
+                        min={1}
+                        max={5}
+                        step={1}
+                        marks
+                        valueLabelDisplay="auto"
+                        valueLabelFormat={(value) => CONFIDENCE_LABELS[value]}
+                    />
+                    <Typography variant="caption" color="text.secondary">
+                        {CONFIDENCE_LABELS[confidence]}
+                    </Typography>
+                </Box>
+
+                {/* Reasoning text area */}
+                <TextField
+                    fullWidth
+                    multiline
+                    rows={4}
+                    label="Reasoning *"
+                    placeholder="Explain why this correction is needed (minimum 10 characters)..."
+                    value={reasoning}
+                    onChange={(e) => {
+                        setReasoning(e.target.value)
+                        setError('')
+                    }}
+                    error={!!error}
+                    helperText={error || `${reasoning.length}/10 minimum characters`}
+                    required
+                />
+            </DialogContent>
+
+            <DialogActions>
+                <Button onClick={handleSkipAndClose} color="inherit">
+                    Skip
+                </Button>
+                <Button onClick={handleSave} variant="contained" color="primary">
+                    Save & Continue
+                </Button>
+            </DialogActions>
+        </Dialog>
+    )
+}
+