| { |
| "$schema": "http://json-schema.org/draft-07/schema#", |
| "$id": "https://github.com/hebrew-unified-nlp/schema/v2.1.0", |
| "title": "HebrewUnifiedNLPResult", |
| "description": "Async parallel Hebrew/English NLP with cross-platform sentence breaking - outputs sentence array with per-sentence language detection", |
| "type": "object", |
| "required": [ |
| "meta", |
| "input", |
| "sentences" |
| ], |
| "properties": { |
| "meta": { |
| "type": "object", |
| "description": "Processing metadata", |
| "required": [ |
| "version", |
| "timestamp", |
| "sentence_count" |
| ], |
| "properties": { |
| "version": { |
| "type": "string", |
| "pattern": "^\\d+\\.\\d+\\.\\d+$", |
| "example": "2025.12.5" |
| }, |
| "timestamp": { |
| "type": "string", |
| "format": "date-time" |
| }, |
| "models": { |
| "type": "object", |
| "properties": { |
| "phonikud": { |
| "type": "string", |
| "description": "Phonikud model name" |
| }, |
| "piper": { |
| "type": "string", |
| "description": "Piper voice manager info" |
| }, |
| "dictabert": { |
| "type": "string", |
| "description": "DictaBERT model path" |
| } |
| } |
| }, |
| "processing_time_ms": { |
| "type": "number", |
| "minimum": 0, |
| "description": "Total processing time for all sentences" |
| }, |
| "sentence_count": { |
| "type": "integer", |
| "minimum": 0, |
| "description": "Number of sentences detected" |
| }, |
| "parallel_workers": { |
| "type": "integer", |
| "minimum": 1, |
| "description": "Number of parallel workers used" |
| }, |
| "sentence_breaker": { |
| "type": "string", |
| "description": "Sentence breaking backend used", |
| "enum": [ |
| "pysbd", |
| "pysbd-en", |
| "icu", |
| "wtpsplit", |
| "regex" |
| ] |
| } |
| } |
| }, |
| "translations": { |
| "type": "object", |
| "description": "Bilingual label translations (Hebrew/English)", |
| "properties": { |
| "pos": { |
| "type": "object", |
| "description": "Part-of-speech translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| }, |
| "dep": { |
| "type": "object", |
| "description": "Dependency relation translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| }, |
| "ner": { |
| "type": "object", |
| "description": "Named entity type translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| }, |
| "prefix": { |
| "type": "object", |
| "description": "Prefix type translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| }, |
| "special": { |
| "type": "object", |
| "description": "Special token translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| }, |
| "morph": { |
| "type": "object", |
| "description": "Morphological feature value translations", |
| "additionalProperties": { |
| "$ref": "#/definitions/BilingualLabel" |
| } |
| } |
| } |
| }, |
| "input": { |
| "type": "object", |
| "required": [ |
| "text" |
| ], |
| "properties": { |
| "text": { |
| "type": "string", |
| "description": "Original input text (may contain multiple sentences)" |
| }, |
| "language": { |
| "type": "string", |
| "description": "Primary input language", |
| "enum": [ |
| "he", |
| "en" |
| ] |
| } |
| } |
| }, |
| "sentences": { |
| "type": "array", |
| "description": "Array of sentence analysis results (parallel processed)", |
| "items": { |
| "$ref": "#/definitions/SentenceResult" |
| } |
| } |
| }, |
| "definitions": { |
| "BilingualLabel": { |
| "type": "object", |
| "description": "Label with Hebrew and English translations", |
| "properties": { |
| "en": { |
| "type": "string" |
| }, |
| "he": { |
| "type": "string" |
| } |
| }, |
| "required": [ |
| "en", |
| "he" |
| ] |
| }, |
| "SentenceResult": { |
| "type": "object", |
| "description": "Analysis result for a single sentence", |
| "required": [ |
| "index", |
| "lang", |
| "text", |
| "phonetics", |
| "tokens" |
| ], |
| "properties": { |
| "index": { |
| "type": "integer", |
| "minimum": 0, |
| "description": "Sentence index in original text" |
| }, |
| "lang": { |
| "type": "string", |
| "enum": [ |
| "he", |
| "en" |
| ], |
| "description": "Detected language for this sentence" |
| }, |
| "text": { |
| "type": "string", |
| "description": "Original sentence text" |
| }, |
| "phonetics": { |
| "$ref": "#/definitions/Phonetics" |
| }, |
| "tokens": { |
| "type": "array", |
| "items": { |
| "$ref": "#/definitions/Token" |
| } |
| }, |
| "ner_entities": { |
| "type": "array", |
| "items": { |
| "$ref": "#/definitions/NEREntity" |
| } |
| }, |
| "speech": { |
| "oneOf": [ |
| { |
| "$ref": "#/definitions/Speech" |
| }, |
| { |
| "type": "null" |
| } |
| ] |
| }, |
| "tree_svg": { |
| "type": "string", |
| "description": "SVG string of dependency tree visualization" |
| } |
| } |
| }, |
| "Phonetics": { |
| "type": "object", |
| "description": "Phonikud output for sentence (Hebrew) or passthrough (English)", |
| "required": [ |
| "diacritized", |
| "phonemes" |
| ], |
| "properties": { |
| "diacritized": { |
| "type": "string", |
| "description": "Sentence with nikud (Hebrew) or original text (English)" |
| }, |
| "phonemes": { |
| "type": "string", |
| "description": "Phoneme sequence (Hebrew) or empty string (English)" |
| } |
| } |
| }, |
| "Token": { |
| "type": "object", |
| "description": "Unified token: DictaBERT + Phonikud enrichment", |
| "required": [ |
| "id", |
| "token" |
| ], |
| "properties": { |
| "id": { |
| "type": "integer", |
| "minimum": 1, |
| "description": "1-indexed token position within sentence" |
| }, |
| "token": { |
| "type": "string", |
| "description": "Surface form (DictaBERT)" |
| }, |
| "nikud": { |
| "type": "string", |
| "description": "Diacritized form (Phonikud for Hebrew, original for English)" |
| }, |
| "phonemes": { |
| "type": "string", |
| "description": "Token phonemes (Hebrew) or empty (English)" |
| }, |
| "seg": { |
| "type": "array", |
| "description": "Morphological segmentation (DictaBERT)", |
| "items": { |
| "type": "string" |
| } |
| }, |
| "lex": { |
| "type": "string", |
| "description": "Lemma (DictaBERT)" |
| }, |
| "offsets": { |
| "type": "object", |
| "properties": { |
| "start": { |
| "type": "integer" |
| }, |
| "end": { |
| "type": "integer" |
| } |
| } |
| }, |
| "morph": { |
| "$ref": "#/definitions/Morphology" |
| }, |
| "syntax": { |
| "$ref": "#/definitions/Syntax" |
| } |
| } |
| }, |
| "Morphology": { |
| "type": "object", |
| "description": "DictaBERT morphological analysis", |
| "properties": { |
| "token": { |
| "type": "string" |
| }, |
| "pos": { |
| "type": "string", |
| "enum": [ |
| "ADJ", |
| "ADP", |
| "ADV", |
| "AUX", |
| "CCONJ", |
| "DET", |
| "INTJ", |
| "NOUN", |
| "NUM", |
| "PART", |
| "PRON", |
| "PROPN", |
| "PUNCT", |
| "SCONJ", |
| "SYM", |
| "VERB", |
| "X" |
| ] |
| }, |
| "feats": { |
| "type": "object", |
| "properties": { |
| "Gender": { |
| "type": "string", |
| "enum": [ |
| "Masc", |
| "Fem" |
| ] |
| }, |
| "Number": { |
| "type": "string", |
| "enum": [ |
| "Sing", |
| "Plur", |
| "Dual" |
| ] |
| }, |
| "Person": { |
| "type": "string", |
| "enum": [ |
| "1", |
| "2", |
| "3" |
| ] |
| }, |
| "Tense": { |
| "type": "string", |
| "enum": [ |
| "Past", |
| "Present", |
| "Future", |
| "Imp", |
| "Inf" |
| ] |
| }, |
| "Voice": { |
| "type": "string", |
| "enum": [ |
| "Act", |
| "Pass" |
| ] |
| }, |
| "Definite": { |
| "type": "string", |
| "enum": [ |
| "Def", |
| "Ind" |
| ] |
| }, |
| "Case": { |
| "type": "string", |
| "enum": [ |
| "Nom", |
| "Acc", |
| "Gen" |
| ] |
| }, |
| "Construct": { |
| "type": "string", |
| "enum": [ |
| "Construct", |
| "Free" |
| ] |
| } |
| }, |
| "additionalProperties": true |
| }, |
| "prefixes": { |
| "type": "array", |
| "items": { |
| "type": "string" |
| } |
| }, |
| "suffix": { |
| "type": "boolean" |
| } |
| } |
| }, |
| "Syntax": { |
| "type": "object", |
| "description": "DictaBERT dependency syntax", |
| "properties": { |
| "word": { |
| "type": "string" |
| }, |
| "dep_head_idx": { |
| "type": "integer", |
| "minimum": -1, |
| "description": "-1 for root" |
| }, |
| "dep_func": { |
| "type": "string" |
| }, |
| "dep_head": { |
| "type": "string" |
| } |
| } |
| }, |
| "NEREntity": { |
| "type": "object", |
| "properties": { |
| "phrase": { |
| "type": "string", |
| "description": "Entity text" |
| }, |
| "label": { |
| "type": "string", |
| "enum": [ |
| "PER", |
| "LOC", |
| "GPE", |
| "ORG", |
| "TIME", |
| "DATE", |
| "MISC", |
| "MONEY", |
| "PERCENT" |
| ] |
| }, |
| "start": { |
| "type": "integer" |
| }, |
| "end": { |
| "type": "integer" |
| } |
| } |
| }, |
| "Speech": { |
| "type": "object", |
| "description": "TTS audio for sentence", |
| "properties": { |
| "format": { |
| "type": "string", |
| "enum": [ |
| "wav", |
| "mp3", |
| "ogg" |
| ] |
| }, |
| "sample_rate": { |
| "type": "integer" |
| }, |
| "duration_ms": { |
| "type": "number" |
| }, |
| "data_uri": { |
| "type": "string", |
| "description": "Base64 data URI (data:audio/wav;base64,...)" |
| } |
| } |
| } |
| } |
| } |
