eric/audita
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity

Implemented deterministic transcript normalization before the LLM stages

Browse Source
This commit is contained in:
Eric Rakestraw
2026-04-22 07:22:40 -05:00
parent 445329de81
commit c1672af9b4
12 changed files with 719 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
tests/test_config.py
View File

@@ -10,6 +10,10 @@ from audita.config import (
DEFAULT_GRAMMAR_MAX_LLM_PASSES,
DEFAULT_MAX_RETRIES,
DEFAULT_MAX_SECTION_TOKENS,
DEFAULT_NORMALIZE_ELLIPSIS_GAP,
DEFAULT_NORMALIZE_MAX_SEGMENT_DURATION,
DEFAULT_NORMALIZE_MAX_SEGMENT_GAP,
DEFAULT_NORMALIZE_MAX_SEGMENT_TOKENS,
DEFAULT_WORK_DIR,
)
from audita.errors import AuditaConfigError
@@ -26,6 +30,14 @@ def test_config_uses_defaults_with_api_key():
assert config.max_retries == DEFAULT_MAX_RETRIES
assert config.glossary_max_llm_passes == DEFAULT_GLOSSARY_MAX_LLM_PASSES
assert config.grammar_max_llm_passes == DEFAULT_GRAMMAR_MAX_LLM_PASSES
assert config.normalize_max_segment_gap == DEFAULT_NORMALIZE_MAX_SEGMENT_GAP
assert config.normalize_max_segment_gap == 5.0
assert config.normalize_ellipsis_gap == DEFAULT_NORMALIZE_ELLIPSIS_GAP
assert config.normalize_ellipsis_gap == 2.0
assert config.normalize_max_segment_duration == DEFAULT_NORMALIZE_MAX_SEGMENT_DURATION
assert config.normalize_max_segment_duration == 60.0
assert config.normalize_max_segment_tokens == DEFAULT_NORMALIZE_MAX_SEGMENT_TOKENS
assert config.normalize_max_segment_tokens == 2048
assert config.work_dir == Path(DEFAULT_WORK_DIR)
@@ -39,6 +51,10 @@ def test_config_env_overrides_defaults():
"AUDITA_MAX_RETRIES": "5",
"AUDITA_GLOSSARY_MAX_LLM_PASSES": "7",
"AUDITA_GRAMMAR_MAX_LLM_PASSES": "4",
"AUDITA_NORMALIZE_MAX_SEGMENT_GAP": "4.5",
"AUDITA_NORMALIZE_ELLIPSIS_GAP": "1.5",
"AUDITA_NORMALIZE_MAX_SEGMENT_DURATION": "45.0",
"AUDITA_NORMALIZE_MAX_SEGMENT_TOKENS": "512",
"AUDITA_WORK_DIR": "/tmp/custom-audita",
}
)
@@ -49,6 +65,10 @@ def test_config_env_overrides_defaults():
assert config.max_retries == 5
assert config.glossary_max_llm_passes == 7
assert config.grammar_max_llm_passes == 4
assert config.normalize_max_segment_gap == 4.5
assert config.normalize_ellipsis_gap == 1.5
assert config.normalize_max_segment_duration == 45.0
assert config.normalize_max_segment_tokens == 512
assert config.work_dir == Path("/tmp/custom-audita")
@@ -60,6 +80,10 @@ def test_config_cli_overrides_env():
"AUDITA_MAX_RETRIES": "5",
"AUDITA_GLOSSARY_MAX_LLM_PASSES": "7",
"AUDITA_GRAMMAR_MAX_LLM_PASSES": "6",
"AUDITA_NORMALIZE_MAX_SEGMENT_GAP": "4.5",
"AUDITA_NORMALIZE_ELLIPSIS_GAP": "1.5",
"AUDITA_NORMALIZE_MAX_SEGMENT_DURATION": "45.0",
"AUDITA_NORMALIZE_MAX_SEGMENT_TOKENS": "512",
"AUDITA_WORK_DIR": "/tmp/env-audita",
},
overrides=ConfigOverrides(
@@ -69,6 +93,10 @@ def test_config_cli_overrides_env():
max_retries=3,
glossary_max_llm_passes=2,
grammar_max_llm_passes=3,
normalize_max_segment_gap=3.0,
normalize_ellipsis_gap=1.0,
normalize_max_segment_duration=30.0,
normalize_max_segment_tokens=256,
work_dir=Path("/tmp/cli-audita"),
),
)
@@ -79,6 +107,10 @@ def test_config_cli_overrides_env():
assert config.max_retries == 3
assert config.glossary_max_llm_passes == 2
assert config.grammar_max_llm_passes == 3
assert config.normalize_max_segment_gap == 3.0
assert config.normalize_ellipsis_gap == 1.0
assert config.normalize_max_segment_duration == 30.0
assert config.normalize_max_segment_tokens == 256
assert config.work_dir == Path("/tmp/cli-audita")
@@ -126,3 +158,20 @@ def test_legacy_confidence_threshold_env_is_ignored():
assert config.glossary_confidence_threshold == DEFAULT_GLOSSARY_CONFIDENCE_THRESHOLD
assert config.grammar_confidence_threshold == DEFAULT_GRAMMAR_CONFIDENCE_THRESHOLD
def test_config_rejects_invalid_normalization_values():
invalid_envs = [
{"AUDITA_NORMALIZE_MAX_SEGMENT_GAP": "-1"},
{"AUDITA_NORMALIZE_ELLIPSIS_GAP": "-1"},
{
"AUDITA_NORMALIZE_MAX_SEGMENT_GAP": "1",
"AUDITA_NORMALIZE_ELLIPSIS_GAP": "2",
},
{"AUDITA_NORMALIZE_MAX_SEGMENT_DURATION": "0"},
{"AUDITA_NORMALIZE_MAX_SEGMENT_TOKENS": "0"},
{"AUDITA_NORMALIZE_MAX_SEGMENT_GAP": "nan"},
]
for env in invalid_envs:
with pytest.raises(AuditaConfigError):
AuditaConfig.from_sources(env={"OPENROUTER_API_KEY": "key", **env})