Implemented support for the seriatim JSON schema for input transcripts

src/audita/core/schemas.py

@@ -8,7 +8,7 @@ from .errors import AuditaValidationError
 
 
 class TranscriptSegment(BaseModel):
-    model_config = ConfigDict(extra="forbid")
+    model_config = ConfigDict(extra="ignore")
 
     id: int = Field(ge=1)
     speaker: StrictStr
@@ -50,7 +50,7 @@ class TranscriptSegment(BaseModel):
 
 
 class SourceTranscriptSegment(BaseModel):
-    model_config = ConfigDict(extra="forbid")
+    model_config = ConfigDict(extra="ignore")
 
     id: Optional[int] = Field(default=None, ge=1)
     speaker: StrictStr
@@ -146,8 +146,9 @@ def validate_transcript_data(
     data: Any,
     require_sequential_ids: bool = True,
 ) -> List[TranscriptSegment]:
+    data = _extract_transcript_segments(data)
     if not isinstance(data, list):
-        raise AuditaValidationError("Transcript must be a JSON array.")
+        raise AuditaValidationError("Transcript must be a JSON array or an object with a segments array.")
     if not data:
         raise AuditaValidationError("Transcript must contain at least one segment.")
     try:
@@ -160,8 +161,9 @@ def validate_transcript_data(
 
 
 def validate_source_transcript_data(data: Any) -> List[SourceTranscriptSegment]:
+    data = _extract_transcript_segments(data)
     if not isinstance(data, list):
-        raise AuditaValidationError("Transcript must be a JSON array.")
+        raise AuditaValidationError("Transcript must be a JSON array or an object with a segments array.")
     if not data:
         raise AuditaValidationError("Transcript must contain at least one segment.")
     try:
@@ -215,6 +217,12 @@ def source_transcript_to_json(segments: List[SourceTranscriptSegment]) -> str:
     return json.dumps(payload, ensure_ascii=False, indent=2) + "\n"
 
 
+def _extract_transcript_segments(data: Any) -> Any:
+    if isinstance(data, dict) and "segments" in data:
+        return data["segments"]
+    return data
+
+
 def _validate_sequential_ids(transcript: List[TranscriptSegment]) -> None:
     ids = [segment.id for segment in transcript]
     expected = list(range(1, len(transcript) + 1))

tests/test_new_transcript_schema.py

@@ -0,0 +1,69 @@
+from audita.core.schemas import parse_source_transcript_json, parse_transcript_json
+
+
+SERIATIM_TRANSCRIPT = """
+{
+  "metadata": {
+    "application": "seriatim",
+    "version": "dev",
+    "input_reader": "json-files",
+    "input_files": ["eric.json", "mike.json"],
+    "preprocessing_modules": ["validate-raw", "normalize-speakers", "trim-text"],
+    "postprocessing_modules": ["detect-overlaps", "resolve-overlaps", "backchannel"],
+    "output_modules": ["json"]
+  },
+  "segments": [
+    {
+      "id": 1,
+      "source": "eric.json",
+      "source_segment_index": 0,
+      "speaker": "Eric Rakestraw",
+      "start": 1.25,
+      "end": 3.5,
+      "text": "Hello there.",
+      "overlap_group_id": 1
+    },
+    {
+      "id": 2,
+      "source": "eric.json",
+      "source_ref": "word-run:1:1:1",
+      "derived_from": ["eric.json#0"],
+      "speaker": "Eric Rakestraw",
+      "start": 4.0,
+      "end": 4.5,
+      "text": "Resolved word run",
+      "categories": ["backchannel"]
+    }
+  ],
+  "overlap_groups": [
+    {
+      "id": 1,
+      "start": 1.25,
+      "end": 4.0,
+      "segments": ["eric.json#0", "mike.json#0"],
+      "speakers": ["Eric Rakestraw", "Mike Brown"],
+      "class": "unknown",
+      "resolution": "unresolved"
+    }
+  ]
+}
+"""
+
+
+def test_parse_source_transcript_json_accepts_seriatim_transcript_object():
+    segments = parse_source_transcript_json(SERIATIM_TRANSCRIPT)
+
+    assert len(segments) == 2
+    assert segments[0].id == 1
+    assert segments[0].speaker == "Eric Rakestraw"
+    assert segments[0].start == 1.25
+    assert segments[0].end == 3.5
+    assert segments[0].text == "Hello there."
+    assert segments[1].text == "Resolved word run"
+
+
+def test_parse_transcript_json_accepts_seriatim_transcript_object_and_ignores_unused_fields():
+    segments = parse_transcript_json(SERIATIM_TRANSCRIPT)
+
+    assert [segment.id for segment in segments] == [1, 2]
+    assert [segment.text for segment in segments] == ["Hello there.", "Resolved word run"]