Files
icassp2027-experiment/test_audit_easycom_floor_events.py
2026-08-30 17:02:13 +08:00

126 lines
4.1 KiB
Python

"""Tests for conservative EasyCom floor-event auditing."""
from __future__ import annotations
import unittest
from audit_easycom_floor_events import derive_eot_candidates
from derive_easycom_interaction_events import Rules, Utterance
def utterance(
uid: str,
participant: int,
start: float,
end: float,
text: str,
*,
cross_clip: bool = False,
starts_at_boundary: bool = False,
ends_at_boundary: bool = False,
) -> Utterance:
return Utterance(
uid=uid,
session="Session_7",
clip_stem="00-00-000",
clip_index=0,
segment_id=0,
participant_id=participant,
target_of_speech=("Group",),
transcription=text,
source_start_frame=round(start * 20) + 1,
source_end_frame=round(end * 20),
dense_start_frame=round(start * 20),
dense_end_frame_exclusive=round(end * 20),
dense_start_seconds=start,
dense_end_seconds=end,
duration_seconds=end - start,
starts_at_clip_boundary=starts_at_boundary,
ends_at_clip_boundary=ends_at_boundary,
annotation_end_frame_clamped_by=0,
is_cross_clip_continuation=cross_clip,
)
class EasyComFloorAuditTests(unittest.TestCase):
def setUp(self) -> None:
self.rules = Rules()
def candidates(self, rows: list[Utterance]):
return derive_eot_candidates(
rows,
camera_wearer_id=2,
rules=self.rules,
)
def test_sustained_overlapping_incoming_speaker_is_positive_proxy(self) -> None:
holder = utterance("holder", 4, 0.0, 2.0, "I finished my point.")
incoming = utterance("incoming", 2, 1.7, 3.2, "Here is my response.")
result = {row.holder_uid: row for row in self.candidates([holder, incoming])}
self.assertEqual(
result["holder"].category,
"eot_positive_overlap_takeover_proxy",
)
self.assertEqual(result["holder"].incoming_uid, "incoming")
def test_same_speaker_resumption_is_mid_turn_pause_negative(self) -> None:
first = utterance("first", 4, 0.0, 1.0, "First clause.")
resumed = utterance("resumed", 4, 1.3, 2.5, "Second clause.")
result = {row.holder_uid: row for row in self.candidates([first, resumed])}
self.assertEqual(
result["first"].category,
"eot_negative_mid_turn_pause_proxy",
)
self.assertAlmostEqual(result["first"].pause_seconds, 0.3)
def test_long_same_speaker_gap_is_ambiguous(self) -> None:
first = utterance("first", 4, 0.0, 1.0, "First clause.")
resumed = utterance("resumed", 4, 4.0, 5.0, "Much later.")
result = {row.holder_uid: row for row in self.candidates([first, resumed])}
self.assertEqual(
result["first"].category,
"eot_ambiguous_long_same_speaker_gap",
)
def test_short_unresolved_incoming_speaker_is_kept_for_review(self) -> None:
holder = utterance("holder", 4, 0.0, 1.0, "First speaker.")
short = utterance("short", 2, 1.2, 1.6, "Maybe.")
resumed = utterance("resumed", 4, 3.5, 4.5, "First speaker continues.")
result = {
row.holder_uid: row
for row in self.candidates([holder, short, resumed])
}
self.assertEqual(
result["holder"].category,
"eot_ambiguous_short_or_unresolved_incoming",
)
self.assertEqual(result["holder"].incoming_uid, "short")
def test_adjacent_cross_file_continuation_is_excluded(self) -> None:
first = utterance(
"first",
4,
0.0,
60.0,
"Continued [X]",
ends_at_boundary=True,
)
resumed = utterance(
"resumed",
4,
60.0,
61.0,
"[X] continuation",
cross_clip=True,
starts_at_boundary=True,
)
result = {row.holder_uid: row for row in self.candidates([first, resumed])}
self.assertEqual(
result["first"].category,
"eot_excluded_file_continuation",
)
if __name__ == "__main__":
unittest.main()