karaoke-gen 0.57.0__py3-none-any.whl → 0.71.23__py3-none-any.whl

lyrics_transcriber/correction/handlers/syllables_match.py

@@ -0,0 +1,252 @@
+from typing import List, Tuple, Dict, Any, Optional
+import spacy
+import logging
+import pyphen
+import nltk
+from nltk.corpus import cmudict
+import syllables
+from spacy_syllables import SpacySyllables
+
+from lyrics_transcriber.types import GapSequence, WordCorrection
+from lyrics_transcriber.correction.handlers.base import GapCorrectionHandler
+from lyrics_transcriber.correction.handlers.word_operations import WordOperations
+
+
+class SyllablesMatchHandler(GapCorrectionHandler):
+    """Handles gaps where number of syllables in reference text matches number of syllables in transcription."""
+
+    def __init__(self, logger: Optional[logging.Logger] = None):
+        super().__init__(logger)
+        self.logger = logger or logging.getLogger(__name__)
+
+        # Marking SpacySyllables as used to prevent unused import warning
+        _ = SpacySyllables
+
+        # Load spacy model with syllables pipeline
+        try:
+            self.nlp = spacy.load("en_core_web_sm")
+        except OSError:
+            self.logger.info("Language model 'en_core_web_sm' not found. Attempting to download...")
+            import subprocess
+
+            try:
+                subprocess.check_call(["python", "-m", "spacy", "download", "en_core_web_sm"])
+                self.nlp = spacy.load("en_core_web_sm")
+                self.logger.info("Successfully downloaded and loaded en_core_web_sm")
+            except subprocess.CalledProcessError as e:
+                raise OSError(
+                    "Language model 'en_core_web_sm' could not be downloaded. "
+                    "Please install it manually with: python -m spacy download en_core_web_sm"
+                ) from e
+
+        # Add syllables component to pipeline if not already present
+        if "syllables" not in self.nlp.pipe_names:
+            self.nlp.add_pipe("syllables", after="tagger")
+
+        # Initialize Pyphen for English
+        self.dic = pyphen.Pyphen(lang="en_US")
+
+        # Initialize NLTK's CMU dictionary
+        try:
+            self.cmudict = cmudict.dict()
+        except LookupError:
+            nltk.download("cmudict")
+            self.cmudict = cmudict.dict()
+
+    def _count_syllables_spacy(self, words: List[str]) -> int:
+        """Count syllables using spacy_syllables."""
+        text = " ".join(words)
+        doc = self.nlp(text)
+        total_syllables = sum(token._.syllables_count or 1 for token in doc)
+        return total_syllables
+
+    def _count_syllables_pyphen(self, words: List[str]) -> int:
+        """Count syllables using pyphen."""
+        total_syllables = 0
+        for word in words:
+            hyphenated = self.dic.inserted(word)
+            syllables_count = len(hyphenated.split("-")) if hyphenated else 1
+            total_syllables += syllables_count
+        return total_syllables
+
+    def _count_syllables_nltk(self, words: List[str]) -> int:
+        """Count syllables using NLTK's CMU dictionary."""
+        total_syllables = 0
+        for word in words:
+            word = word.lower()
+            if word in self.cmudict:
+                syllables_count = len([ph for ph in self.cmudict[word][0] if ph[-1].isdigit()])
+                total_syllables += syllables_count
+            else:
+                total_syllables += 1
+        return total_syllables
+
+    def _count_syllables_lib(self, words: List[str]) -> int:
+        """Count syllables using the syllables library."""
+        total_syllables = 0
+        for word in words:
+            syllables_count = syllables.estimate(word)
+            total_syllables += syllables_count
+        return total_syllables
+
+    def _count_syllables(self, words: List[str]) -> List[int]:
+        """Count syllables using multiple methods."""
+        spacy_count = self._count_syllables_spacy(words)
+        pyphen_count = self._count_syllables_pyphen(words)
+        nltk_count = self._count_syllables_nltk(words)
+        syllables_count = self._count_syllables_lib(words)
+
+        text = " ".join(words)
+        self.logger.debug(
+            f"Syllable counts for '{text}': spacy={spacy_count}, pyphen={pyphen_count}, nltk={nltk_count}, syllables={syllables_count}"
+        )
+        return [spacy_count, pyphen_count, nltk_count, syllables_count]
+
+    def can_handle(self, gap: GapSequence, data: Optional[Dict[str, Any]] = None) -> Tuple[bool, Dict[str, Any]]:
+        # Must have reference words
+        if not gap.reference_word_ids:
+            self.logger.debug("No reference word IDs available")
+            return False, {}
+
+        # Get word lookup map from data
+        if not data or "word_map" not in data:
+            self.logger.error("No word_map provided in data")
+            return False, {}
+
+        word_map = data["word_map"]
+
+        # Get actual words from word IDs
+        gap_words = []
+        for word_id in gap.transcribed_word_ids:
+            if word_id not in word_map:
+                self.logger.error(f"Word ID {word_id} not found in word_map")
+                return False, {}
+            gap_words.append(word_map[word_id].text)
+
+        # Get syllable counts for gap text using different methods
+        gap_syllables = self._count_syllables(gap_words)
+
+        # Check if any reference source has matching syllable count with any method
+        for source, ref_word_ids in gap.reference_word_ids.items():
+            # Get reference words from word map
+            ref_words = []
+            for word_id in ref_word_ids:
+                if word_id not in word_map:
+                    self.logger.error(f"Reference word ID {word_id} not found in word_map")
+                    continue
+                ref_words.append(word_map[word_id].text)
+
+            if not ref_words:
+                continue
+
+            ref_syllables = self._count_syllables(ref_words)
+
+            # If any counting method matches between gap and reference, we can handle it
+            if any(gap_count == ref_count for gap_count in gap_syllables for ref_count in ref_syllables):
+                self.logger.debug(f"Found matching syllable count in source '{source}'")
+                return True, {
+                    "gap_syllables": gap_syllables,
+                    "matching_source": source,
+                    "reference_word_ids": ref_word_ids,
+                    "word_map": word_map,
+                }
+
+        self.logger.debug("No reference source had matching syllable count")
+        return False, {}
+
+    def handle(self, gap: GapSequence, data: Optional[Dict[str, Any]] = None) -> List[WordCorrection]:
+        """Handle the gap using syllable matching."""
+        if not data:
+            can_handle, data = self.can_handle(gap)
+            if not can_handle:
+                return []
+
+        corrections = []
+        matching_source = data["matching_source"]
+        reference_word_ids = data["reference_word_ids"]
+        word_map = data["word_map"]
+
+        # Get the actual words from word IDs
+        gap_words = [word_map[word_id].text for word_id in gap.transcribed_word_ids]
+        ref_words = [word_map[word_id].text for word_id in reference_word_ids]
+
+        # Use the centralized method to calculate reference positions
+        reference_positions = WordOperations.calculate_reference_positions(gap, [matching_source])
+
+        # Since we matched syllable counts for the entire gap, we should handle all words
+        if len(gap_words) > len(ref_words):
+            # Multiple transcribed words -> fewer reference words
+            # Try to distribute the reference words across the gap words
+            words_per_ref = len(gap_words) / len(ref_words)
+
+            for ref_idx, ref_word_id in enumerate(reference_word_ids):
+                start_idx = int(ref_idx * words_per_ref)
+                end_idx = int((ref_idx + 1) * words_per_ref)
+
+                # Get the group of words to combine
+                words_to_combine = gap_words[start_idx:end_idx]
+                word_ids_to_combine = gap.transcribed_word_ids[start_idx:end_idx]
+                corrections.extend(
+                    WordOperations.create_word_combine_corrections(
+                        original_words=words_to_combine,
+                        reference_word=word_map[ref_word_id].text,
+                        original_position=gap.transcription_position + start_idx,
+                        source=matching_source,
+                        confidence=0.8,
+                        combine_reason="Words combined based on syllable match",
+                        delete_reason="Word removed as part of syllable match combination",
+                        reference_positions=reference_positions,
+                        handler="SyllablesMatchHandler",
+                        original_word_ids=word_ids_to_combine,
+                        corrected_word_id=ref_word_id,
+                    )
+                )
+
+        elif len(gap_words) < len(ref_words):
+            # Single transcribed word -> multiple reference words
+            words_per_gap = len(ref_words) / len(gap_words)
+
+            for i, word_id in enumerate(gap.transcribed_word_ids):
+                start_idx = int(i * words_per_gap)
+                end_idx = int((i + 1) * words_per_gap)
+                ref_word_ids_for_split = reference_word_ids[start_idx:end_idx]
+                ref_words_for_split = [word_map[ref_id].text for ref_id in ref_word_ids_for_split]
+
+                corrections.extend(
+                    WordOperations.create_word_split_corrections(
+                        original_word=word_map[word_id].text,
+                        reference_words=ref_words_for_split,
+                        original_position=gap.transcription_position + i,
+                        source=matching_source,
+                        confidence=0.8,
+                        reason="Split word based on syllable match",
+                        reference_positions=reference_positions,
+                        handler="SyllablesMatchHandler",
+                        original_word_id=word_id,
+                        corrected_word_ids=ref_word_ids_for_split,
+                    )
+                )
+
+        else:
+            # One-to-one replacement
+            for i, (orig_word_id, ref_word_id) in enumerate(zip(gap.transcribed_word_ids, reference_word_ids)):
+                orig_word = word_map[orig_word_id]
+                ref_word = word_map[ref_word_id]
+
+                if orig_word.text.lower() != ref_word.text.lower():
+                    corrections.append(
+                        WordOperations.create_word_replacement_correction(
+                            original_word=orig_word.text,
+                            corrected_word=ref_word.text,
+                            original_position=gap.transcription_position + i,
+                            source=matching_source,
+                            confidence=0.8,
+                            reason=f"Source '{matching_source}' had matching syllable count",
+                            reference_positions=reference_positions,
+                            handler="SyllablesMatchHandler",
+                            original_word_id=orig_word_id,
+                            corrected_word_id=ref_word_id,
+                        )
+                    )
+
+        return corrections

lyrics_transcriber/correction/handlers/word_count_match.py

@@ -0,0 +1,80 @@
+from typing import List, Tuple, Dict, Any, Optional
+import logging
+
+from lyrics_transcriber.types import GapSequence, WordCorrection
+from lyrics_transcriber.correction.handlers.base import GapCorrectionHandler
+from lyrics_transcriber.correction.handlers.word_operations import WordOperations
+
+
+class WordCountMatchHandler(GapCorrectionHandler):
+    """Handles gaps where reference sources agree and have matching word counts."""
+
+    def __init__(self, logger: Optional[logging.Logger] = None):
+        super().__init__(logger)
+        self.logger = logger or logging.getLogger(__name__)
+
+    def can_handle(self, gap: GapSequence, data: Optional[Dict[str, Any]] = None) -> Tuple[bool, Dict[str, Any]]:
+        # Must have reference words
+        if not gap.reference_word_ids:
+            self.logger.debug("No reference word IDs available.")
+            return False, {}
+
+        if not self._validate_data(data):
+            return False, {}
+
+        ref_word_lists = list(gap.reference_word_ids.values())
+
+        # All sources must have same number of words as gap
+        if not all(len(words) == gap.length for words in ref_word_lists):
+            self.logger.debug("Not all sources have the same number of words as the gap.")
+            return False, {}
+
+        # If we have multiple sources, they must all agree
+        if len(ref_word_lists) > 1 and not all(words == ref_word_lists[0] for words in ref_word_lists[1:]):
+            self.logger.debug("Not all sources agree on the words.")
+            return False, {}
+
+        self.logger.debug("All sources agree and have matching word counts.")
+        return True, {"word_map": data["word_map"]}
+
+    def handle(self, gap: GapSequence, data: Optional[Dict[str, Any]] = None) -> List[WordCorrection]:
+        if not self._validate_data(data):
+            return []
+
+        corrections = []
+        word_map = data["word_map"]
+        source = list(gap.reference_word_ids.keys())[0]
+        reference_word_ids = gap.reference_word_ids[source]
+        sources = ", ".join(gap.reference_word_ids.keys())
+
+        reference_positions = WordOperations.calculate_reference_positions(gap)
+
+        for i, (orig_word_id, ref_word_id) in enumerate(zip(gap.transcribed_word_ids, reference_word_ids)):
+            # Get the actual words from the word map
+            if orig_word_id not in word_map:
+                self.logger.error(f"Original word ID {orig_word_id} not found in word_map")
+                continue
+            orig_word = word_map[orig_word_id]
+
+            if ref_word_id not in word_map:
+                self.logger.error(f"Reference word ID {ref_word_id} not found in word_map")
+                continue
+            ref_word = word_map[ref_word_id]
+
+            if orig_word.text.lower() != ref_word.text.lower():
+                correction = WordOperations.create_word_replacement_correction(
+                    original_word=orig_word.text,
+                    corrected_word=ref_word.text,
+                    original_position=gap.transcription_position + i,
+                    source=sources,
+                    confidence=1.0,
+                    reason="Reference sources had same word count as gap",
+                    reference_positions=reference_positions,
+                    handler="WordCountMatchHandler",
+                    original_word_id=orig_word_id,
+                    corrected_word_id=ref_word_id,  # Use the reference word's ID
+                )
+                corrections.append(correction)
+                self.logger.debug(f"Correction made: {correction}")
+
+        return corrections

lyrics_transcriber/correction/handlers/word_operations.py

@@ -0,0 +1,187 @@
+from typing import List, Optional, Dict, Any
+from lyrics_transcriber.types import WordCorrection, GapSequence
+from lyrics_transcriber.utils.word_utils import WordUtils
+
+
+class WordOperations:
+    """Utility class for common word manipulation operations used by correction handlers."""
+
+    @staticmethod
+    def calculate_reference_positions(
+        gap: GapSequence, sources: Optional[List[str]] = None, anchor_sequences: Optional[List[Any]] = None
+    ) -> Dict[str, int]:
+        """Calculate reference positions for given sources based on preceding anchor.
+
+        Args:
+            gap: The gap sequence containing the preceding anchor ID
+            sources: Optional list of sources to calculate positions for. If None, uses all sources.
+            anchor_sequences: List of anchor sequences to look up preceding anchor
+
+        Returns:
+            Dictionary mapping source names to their reference positions
+        """
+        reference_positions = {}
+
+        if not gap.preceding_anchor_id or not anchor_sequences:
+            return reference_positions
+
+        # Find the preceding anchor in the sequences
+        preceding_anchor = next(
+            (scored_anchor.anchor for scored_anchor in anchor_sequences if scored_anchor.anchor.id == gap.preceding_anchor_id), None
+        )
+
+        if not preceding_anchor:
+            return reference_positions
+
+        # If no sources specified, use all sources from reference words
+        sources_to_check = sources or list(gap.reference_word_ids.keys())
+
+        for source in sources_to_check:
+            # Get reference positions from the anchor
+            if source in preceding_anchor.reference_positions:
+                # Calculate base position from anchor
+                anchor_pos = preceding_anchor.reference_positions[source]
+                base_ref_pos = anchor_pos + len(preceding_anchor.reference_word_ids[source])
+
+                # Calculate word offset within the gap
+                word_offset = 0
+
+                # Add word offset to base position
+                ref_pos = base_ref_pos + word_offset
+                reference_positions[source] = ref_pos
+
+        return reference_positions
+
+    @staticmethod
+    def create_word_replacement_correction(
+        original_word: str,
+        corrected_word: str,
+        original_position: int,
+        source: str,
+        confidence: float,
+        reason: str,
+        handler: str,
+        reference_positions: Optional[Dict[str, int]] = None,
+        original_word_id: Optional[str] = None,
+        corrected_word_id: Optional[str] = None,
+    ) -> WordCorrection:
+        """Creates a correction for replacing a single word with another word."""
+        return WordCorrection(
+            original_word=original_word,
+            corrected_word=corrected_word,
+            segment_index=0,
+            original_position=original_position,
+            confidence=confidence,
+            source=source,
+            reason=reason,
+            alternatives={},
+            reference_positions=reference_positions,
+            length=1,
+            handler=handler,
+            word_id=original_word_id,
+            corrected_word_id=corrected_word_id if corrected_word_id is not None else (WordUtils.generate_id() if corrected_word else None),
+        )
+
+    @staticmethod
+    def create_word_split_corrections(
+        original_word: str,
+        reference_words: List[str],
+        original_position: int,
+        source: str,
+        confidence: float,
+        reason: str,
+        handler: str,
+        reference_positions: Optional[Dict[str, int]] = None,
+        original_word_id: Optional[str] = None,
+        corrected_word_ids: Optional[List[str]] = None,
+    ) -> List[WordCorrection]:
+        """Creates corrections for splitting a single word into multiple words."""
+        corrections = []
+
+        # Generate word IDs if none provided
+        if corrected_word_ids is None:
+            corrected_word_ids = [WordUtils.generate_id() for _ in reference_words]
+
+        for split_idx, (ref_word, word_id) in enumerate(zip(reference_words, corrected_word_ids)):
+            corrections.append(
+                WordCorrection(
+                    original_word=original_word,
+                    corrected_word=ref_word,
+                    segment_index=0,
+                    original_position=original_position,
+                    confidence=confidence,
+                    source=source,
+                    reason=reason,
+                    alternatives={},
+                    split_index=split_idx,
+                    split_total=len(reference_words),
+                    reference_positions=reference_positions,
+                    length=1,  # Each split word is length 1
+                    handler=handler,
+                    word_id=WordUtils.generate_id(),  # Generate new ID for each split
+                    corrected_word_id=word_id,
+                )
+            )
+        return corrections
+
+    @staticmethod
+    def create_word_combine_corrections(
+        original_words: List[str],
+        reference_word: str,
+        original_position: int,
+        source: str,
+        confidence: float,
+        combine_reason: str,
+        delete_reason: str,
+        handler: str,
+        reference_positions: Optional[Dict[str, int]] = None,
+        original_word_ids: Optional[List[str]] = None,
+        corrected_word_id: Optional[str] = None,
+    ) -> List[WordCorrection]:
+        """Creates corrections for combining multiple words into a single word."""
+        corrections = []
+        word_ids = original_word_ids or [None] * len(original_words)
+
+        final_word_id = corrected_word_id or WordUtils.generate_id()
+
+        # First word gets replaced
+        corrections.append(
+            WordCorrection(
+                original_word=original_words[0],
+                corrected_word=reference_word,
+                segment_index=0,
+                original_position=original_position,
+                confidence=confidence,
+                source=source,
+                reason=combine_reason,
+                alternatives={},
+                reference_positions=reference_positions,
+                length=len(original_words),  # Combined word spans all original words
+                handler=handler,
+                word_id=WordUtils.generate_id(),  # Generate new ID for combined word
+                corrected_word_id=final_word_id,
+            )
+        )
+
+        # Additional words get marked for deletion
+        for i, (word, word_id) in enumerate(zip(original_words[1:], word_ids[1:]), start=1):
+            corrections.append(
+                WordCorrection(
+                    original_word=word,
+                    corrected_word="",
+                    segment_index=0,
+                    original_position=original_position + i,
+                    confidence=confidence,
+                    source=source,
+                    reason=delete_reason,
+                    alternatives={},
+                    is_deletion=True,
+                    reference_positions=reference_positions,
+                    length=1,  # Deleted words are length 1
+                    handler=handler,
+                    word_id=WordUtils.generate_id(),  # Generate new ID for each deleted word
+                    corrected_word_id=None,  # Deleted words don't need a corrected ID
+                )
+            )
+
+        return corrections