eric/seriatim
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 4 Wiki Activity

Updated DefaultWordRunReorderWindow to 1.0

Browse Source
This commit is contained in:
Eric Rakestraw
2026-04-28 15:09:43 -05:00
parent 9cca88280f
commit 47b6727973
6 changed files with 167 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
internal/overlap/resolve.go
View File

@@ -171,7 +171,7 @@ func resolveGroup(in model.MergedTranscript, group model.OverlapGroup, refToInde
speakers := groupSpeakerOrder(group, segmentsBySpeaker)
resolved := resolvedGroup{}
for speakerIndex, speaker := range speakers {
timedWords, untimedWords := gatherResolutionWords(segmentsBySpeaker[speaker], expandedStart, expandedEnd)
timedWords, untimedWords := gatherResolutionWords(segmentsBySpeaker[speaker])
if len(timedWords) == 0 {
continue
}
@@ -303,7 +303,7 @@ func groupSpeakerOrder(group model.OverlapGroup, segmentsBySpeaker map[string][]
return speakers
}
func gatherResolutionWords(segments []model.Segment, groupStart float64, groupEnd float64) ([]resolutionWord, []resolutionWord) {
func gatherResolutionWords(segments []model.Segment) ([]resolutionWord, []resolutionWord) {
timedWords := make([]resolutionWord, 0)
untimedWords := make([]resolutionWord, 0)
sequence := 0
@@ -321,9 +321,6 @@ func gatherResolutionWords(segments []model.Segment, groupStart float64, groupEn
untimedWords = append(untimedWords, candidate)
continue
}
if word.End <= groupStart || word.Start >= groupEnd {
continue
}
timedWords = append(timedWords, candidate)
}
}

91
internal/overlap/resolve_test.go
View File

@@ -72,7 +72,7 @@ func TestResolveCreatesChronologicalWordRunSegments(t *testing.T) {
}
}
func TestResolveIncludesWordsByIntervalIntersection(t *testing.T) {
func TestResolvePreservesAllWordsFromSelectedOverlapSegment(t *testing.T) {
merged := model.MergedTranscript{
Segments: []model.Segment{
segmentWithWords(
@@ -101,11 +101,98 @@ func TestResolveIncludesWordsByIntervalIntersection(t *testing.T) {
if len(got.Segments) != 1 {
t.Fatalf("segment count = %d, want 1", len(got.Segments))
}
if got.Segments[0].Text != "left-edge inside right-edge" {
if got.Segments[0].Text != "before left-edge inside right-edge after" {
t.Fatalf("text = %q", got.Segments[0].Text)
}
}
func TestResolvePreservesPrefixWordsFromAbsorbedContextSegment(t *testing.T) {
merged := model.MergedTranscript{
Segments: []model.Segment{
segmentWithWords(
"a.json",
0,
"Alice",
7,
9.5,
word("prefix", 7.2, 7.4),
word("near", 9.2, 9.4),
),
segmentWithWords("a.json", 1, "Alice", 10, 12, word("inside", 10.5, 10.7)),
segmentWithWords("b.json", 0, "Bob", 10.2, 11.2, word("bob", 10.4, 10.6)),
},
OverlapGroups: []model.OverlapGroup{
group(1, 10, 12, []string{"a.json#1", "b.json#0"}, []string{"Alice", "Bob"}),
},
}
got, _, err := Resolve(merged, 10, 0.4, 3)
if err != nil {
t.Fatalf("resolve failed: %v", err)
}
if gotTexts(got.Segments) != "prefix near inside,bob" {
t.Fatalf("segment texts = %s", gotTexts(got.Segments))
}
}
func TestResolvePreservesSuffixWordsFromAbsorbedContextSegment(t *testing.T) {
merged := model.MergedTranscript{
Segments: []model.Segment{
segmentWithWords("a.json", 0, "Alice", 10, 12, word("inside", 10.5, 10.7)),
segmentWithWords(
"a.json",
1,
"Alice",
12.5,
16,
word("near", 12.7, 12.9),
word("suffix", 15.6, 15.8),
),
segmentWithWords("b.json", 0, "Bob", 10.2, 11.2, word("bob", 10.4, 10.6)),
},
OverlapGroups: []model.OverlapGroup{
group(1, 10, 12, []string{"a.json#0", "b.json#0"}, []string{"Alice", "Bob"}),
},
}
got, _, err := Resolve(merged, 10, 0.4, 3)
if err != nil {
t.Fatalf("resolve failed: %v", err)
}
if gotTexts(got.Segments) != "bob,inside near suffix" {
t.Fatalf("segment texts = %s", gotTexts(got.Segments))
}
}
func TestResolveOutOfWindowWordsAffectWordRunBoundaries(t *testing.T) {
merged := model.MergedTranscript{
Segments: []model.Segment{
segmentWithWords(
"a.json",
0,
"Alice",
7,
9.5,
word("early", 7.2, 7.4),
word("near", 9.2, 9.4),
),
segmentWithWords("a.json", 1, "Alice", 10, 12, word("inside", 10.5, 10.7)),
segmentWithWords("b.json", 0, "Bob", 10.2, 11.2, word("bob", 10.4, 10.6)),
},
OverlapGroups: []model.OverlapGroup{
group(1, 10, 12, []string{"a.json#1", "b.json#0"}, []string{"Alice", "Bob"}),
},
}
got, _, err := Resolve(merged, 0.75, 0.4, 3)
if err != nil {
t.Fatalf("resolve failed: %v", err)
}
if gotTexts(got.Segments) != "early,near,bob,inside" {
t.Fatalf("segment texts = %s", gotTexts(got.Segments))
}
}
func TestResolveIncludesContextWordsAroundOverlapWindow(t *testing.T) {
merged := model.MergedTranscript{
Segments: []model.Segment{