Updated .gitignore to ignore .codex and related files

.gitignore

@@ -1,3 +1,7 @@
+# ---> Codex
+.codex
+AGENTS.md
+
 # ---> Go
 # If you prefer the allow list template instead of the deny list, see community template:
 # https://github.com/github/gitignore/blob/main/community/Golang/Go.AllowList.gitignore

.woodpecker/release.yml

@@ -42,8 +42,9 @@ steps:
         from_secret: GITEA_RELEASE_TOKEN
       files:
         - dist/seriatim-*
-      checksum: dist/seriatim-*
+      checksum: sha256
       checksum-file: SHA256SUMS
+      checksum-flatten: true
       file-exists: skip
       overwrite: false
       prerelease: false

README.md

@@ -49,7 +49,7 @@ Global flags:
 | `--autocorrect` | No | none | Autocorrect rules YAML file. When omitted, the default `autocorrect` module leaves text unchanged. |
 | `--input-reader` | No | `json-files` | Input reader module. |
 | `--output-modules` | No | `json` | Comma-separated output modules. |
-| `--output-schema` | No | `default` | JSON output contract. Allowed values are `default`, `minimal`, and `seriatim`. |
+| `--output-schema` | No | `seriatim-intermediate` | JSON output contract. Allowed values are `seriatim-minimal`, `seriatim-intermediate`, and `seriatim-full`. If omitted, the runtime default is used; consumers that depend on a specific shape should set this explicitly. |
 | `--preprocessing-modules` | No | `validate-raw,normalize-speakers,trim-text` | Comma-separated preprocessing modules, evaluated in order. |
 | `--postprocessing-modules` | No | `detect-overlaps,resolve-overlaps,backchannel,filler,resolve-danglers,coalesce,detect-overlaps,autocorrect,assign-ids,validate-output` | Comma-separated postprocessing modules, evaluated in order. |
 | `--coalesce-gap` | No | `3.0` | Maximum same-speaker gap in seconds for `coalesce`; also used as the `resolve-overlaps` context window. Must be a non-negative float. |
@@ -58,7 +58,8 @@ Environment variables:
 
 | Environment Variable | Default | Description |
 | --- | --- | --- |
-| `SERIATIM_OVERLAP_WORD_RUN_GAP` | `0.75` | Maximum gap in seconds between adjacent timed words when `resolve-overlaps` builds word-run replacement segments. Must be a positive float. |
+| `SERIATIM_OUTPUT_SCHEMA` | `seriatim-intermediate` | Output schema used when `--output-schema` is not explicitly provided. Allowed values are `seriatim-minimal`, `seriatim-intermediate`, and `seriatim-full`. The CLI flag takes precedence. |
+| `SERIATIM_OVERLAP_WORD_RUN_GAP` | `1.0` | Maximum gap in seconds between adjacent timed words when `resolve-overlaps` builds word-run replacement segments. Must be a positive float. |
 | `SERIATIM_OVERLAP_WORD_RUN_REORDER_WINDOW` | `1.0` | Near-start window in seconds for ordering replacement word runs shortest-first. Must be a positive float. |
 | `SERIATIM_BACKCHANNEL_MAX_DURATION` | `2.0` | Maximum duration in seconds for `backchannel` classification. Must be a positive float. |
 | `SERIATIM_FILLER_MAX_DURATION` | `1.25` | Maximum duration in seconds for `filler` classification. Must be a positive float. |
@@ -159,14 +160,16 @@ The old `inputs:` direct mapping format is no longer supported.
 
 `--output-modules json` controls the writer. `--output-schema` controls the JSON contract that writer serializes.
 
-The `default` schema is the default output contract. It stays close to `minimal`, but adds optional `categories` on each segment:
+The named schemas are stable public contracts. If a consumer depends on a specific shape, it should request that schema explicitly at runtime. The runtime default selection may change in a future release.
+
+The `seriatim-intermediate` schema is the current default selection when neither `--output-schema` nor `SERIATIM_OUTPUT_SCHEMA` is set. It stays close to the minimal schema, but adds optional `categories` on each segment:
 
 ```json
 {
   "metadata": {
     "application": "seriatim",
     "version": "dev",
-    "output_schema": "default"
+    "output_schema": "seriatim-intermediate"
   },
   "segments": [
     {
@@ -181,7 +184,7 @@ The `default` schema is the default output contract. It stays close to `minimal`
 }
 ```
 
-The explicit `seriatim` schema uses the full seriatim envelope:
+The `seriatim-full` schema uses the full seriatim envelope:
 
 ```json
 {
@@ -191,7 +194,7 @@ The explicit `seriatim` schema uses the full seriatim envelope:
     "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", "filler", "coalesce", "resolve-danglers", "detect-overlaps", "autocorrect", "assign-ids", "validate-output"],
+    "postprocessing_modules": ["detect-overlaps", "resolve-overlaps", "backchannel", "filler", "resolve-danglers", "coalesce", "detect-overlaps", "autocorrect", "assign-ids", "validate-output"],
     "output_modules": ["json"]
   },
   "segments": [
@@ -231,14 +234,14 @@ The explicit `seriatim` schema uses the full seriatim envelope:
 }
 ```
 
-The `minimal` schema emits minimal metadata and compact ordered segments:
+The `seriatim-minimal` schema emits minimal metadata and compact ordered segments:
 
 ```json
 {
   "metadata": {
     "application": "seriatim",
     "version": "dev",
-    "output_schema": "minimal"
+    "output_schema": "seriatim-minimal"
   },
   "segments": [
     {
@@ -254,7 +257,7 @@ The `minimal` schema emits minimal metadata and compact ordered segments:
 
 Minimal output intentionally omits categories, overlap groups, source/provenance fields, and pipeline configuration metadata.
 
-Default output intentionally omits overlap groups and source/provenance fields, but keeps optional `categories` and minimal metadata.
+Intermediate output intentionally omits overlap groups and source/provenance fields, but keeps optional `categories` and minimal metadata.
 
 Segments are sorted deterministically by:
 
@@ -270,7 +273,7 @@ The public Go output contract is available from:
 import "gitea.maximumdirect.net/eric/seriatim/schema"
 ```
 
-The same package embeds machine-readable JSON Schemas in `schema/output.schema.json`, `schema/default-output.schema.json`, and `schema/minimal-output.schema.json`. The default `validate-output` postprocessor validates the selected output shape and verifies final segment IDs are present, sequential, and start at `1`.
+The same package embeds machine-readable JSON Schemas in `schema/full-output.schema.json`, `schema/intermediate-output.schema.json`, and `schema/minimal-output.schema.json`. The default `validate-output` postprocessor validates the selected output shape and verifies final segment IDs are present, sequential, and start at `1`.
 
 ## Overlap Detection
 
@@ -295,12 +298,12 @@ For each detected overlap group, `resolve-overlaps` uses preserved WhisperX word
 
 - The resolution window expands the detected overlap group by `--coalesce-gap` seconds on both sides.
 - Nearby same-speaker context segments are included when they intersect the expanded window and their start or end is within `--coalesce-gap` of the original overlap boundary.
-- Words are included when their interval intersects the expanded resolution window.
+- Once a segment is selected for replacement, all timed words from that segment participate in word-run construction; the window controls segment selection, not per-word clipping.
 - Context segments that are part of another detected overlap group are not pulled into the current group.
 - Untimed words are included in replacement text in original word order when nearby timed words create a replacement run.
 - Untimed words do not affect replacement segment start/end times or word-run gap splitting.
 - Words for the same speaker are merged into one run when the gap between adjacent words is no greater than `SERIATIM_OVERLAP_WORD_RUN_GAP`.
-- The default word-run gap is `0.75` seconds.
+- The default word-run gap is `1.0` seconds.
 - Set `SERIATIM_OVERLAP_WORD_RUN_GAP` to a positive number of seconds to override the default.
 - Near-start replacement word runs are reordered so shorter segments come first when adjacent starts are within `SERIATIM_OVERLAP_WORD_RUN_REORDER_WINDOW`.
 - The default word-run reorder window is `1.0` seconds.
@@ -339,12 +342,12 @@ The default pipeline runs `resolve-danglers` before `coalesce` and before the se
 
 - Dangling-end fragments have no more than two words and end in punctuation.
 - Dangling-start fragments have no more than two words.
-- Matching uses any shared `derived_from` value.
+- Matching uses same-speaker segments with any shared `derived_from` value.
 - Merged segments use `source_ref` values such as `resolve-danglers:1`, keep the target segment's transcript position, and union `derived_from`.
 
 ## Coalescing
 
-The default pipeline runs `coalesce` after `resolve-danglers` and the second overlap detection pass. It merges adjacent same-speaker segments in the transcript's current order when `next.start - current.end <= --coalesce-gap`.
+The default pipeline runs `coalesce` after `resolve-danglers` and before the second overlap detection pass. It merges adjacent same-speaker segments in the transcript's current order when `next.start - current.end <= --coalesce-gap`.
 
 Coalesced segments use `source_ref` values such as `coalesce:1`, include `derived_from`, and omit `source_segment_index`.
 

architecture.md

@@ -173,6 +173,14 @@ The current output format is JSON, specified with:
 --output-file merged.json
 ```
 
+The current named JSON schemas are:
+
+- `seriatim-minimal`
+- `seriatim-intermediate`
+- `seriatim-full`
+
+The current runtime default selection is `seriatim-intermediate`, but default selection may change over time. Consumers that depend on a specific schema should request it explicitly.
+
 Future output formats may include:
 
 - Markdown.
@@ -216,7 +224,7 @@ seriatim merge \
   --preprocessing-modules validate-raw,normalize-speakers,trim-text \
   --postprocessing-modules detect-overlaps,resolve-overlaps,backchannel,filler,resolve-danglers,coalesce,detect-overlaps,autocorrect,assign-ids,validate-output \
   --output-modules json \
-  --output-schema default \
+  --output-schema seriatim-intermediate \
   --output-file merged.json \
   --report-file report.json
 ```
@@ -358,7 +366,7 @@ Initial classifications may include:
 - `backchannel`
 - `crosstalk`
 
-The `resolve-overlaps` module uses preserved word-level timing to replace detected overlap-group segments with smaller word-run segments when usable timing is available. Resolution expands each overlap window by the configured coalesce gap so nearby same-speaker context can be absorbed into the replacement runs. Groups without usable word timing remain unresolved for later passes or human review.
+The `resolve-overlaps` module uses preserved word-level timing to replace detected overlap-group segments with smaller word-run segments when usable timing is available. Resolution expands each overlap window by the configured coalesce gap so nearby same-speaker context can be absorbed into the replacement runs. Once a segment is selected for replacement, all timed words from that segment participate in word-run construction so text is not clipped at the window boundary. Groups without usable word timing remain unresolved for later passes or human review.
 
 Overlap resolution should be non-destructive. Original segment text, timing, and source metadata must remain recoverable.
 

internal/artifact/transcript.go

@@ -57,12 +57,12 @@ func FromMerged(cfg config.Config, merged model.MergedTranscript) schema.Transcr
 	}
 }
 
-// DefaultFromMerged converts the internal merged transcript model into the
+// IntermediateFromMerged converts the internal merged transcript model into the
-// compact default public serialized output contract.
+// compact intermediate public serialized output contract.
-func DefaultFromMerged(cfg config.Config, merged model.MergedTranscript) schema.DefaultTranscript {
+func IntermediateFromMerged(cfg config.Config, merged model.MergedTranscript) schema.IntermediateTranscript {
-	segments := make([]schema.DefaultSegment, len(merged.Segments))
+	segments := make([]schema.IntermediateSegment, len(merged.Segments))
 	for index, segment := range merged.Segments {
-		segments[index] = schema.DefaultSegment{
+		segments[index] = schema.IntermediateSegment{
 			ID:         segment.ID,
 			Start:      segment.Start,
 			End:        segment.End,
@@ -72,11 +72,11 @@ func DefaultFromMerged(cfg config.Config, merged model.MergedTranscript) schema.
 		}
 	}
 
-	return schema.DefaultTranscript{
+	return schema.IntermediateTranscript{
-		Metadata: schema.DefaultMetadata{
+		Metadata: schema.IntermediateMetadata{
 			Application:  ApplicationName,
 			Version:      buildinfo.Version,
-			OutputSchema: config.OutputSchemaDefault,
+			OutputSchema: config.OutputSchemaIntermediate,
 		},
 		Segments: segments,
 	}
@@ -110,14 +110,14 @@ func MinimalFromMerged(cfg config.Config, merged model.MergedTranscript) schema.
 // runtime-selected public output contract.
 func SelectedFromMerged(cfg config.Config, merged model.MergedTranscript) any {
 	switch cfg.OutputSchema {
-	case config.OutputSchemaDefault:
-		return DefaultFromMerged(cfg, merged)
 	case config.OutputSchemaMinimal:
 		return MinimalFromMerged(cfg, merged)
-	case config.OutputSchemaSeriatim:
+	case config.OutputSchemaFull:
 		return FromMerged(cfg, merged)
+	case config.OutputSchemaIntermediate:
+		return IntermediateFromMerged(cfg, merged)
 	default:
-		return FromMerged(cfg, merged)
+		return IntermediateFromMerged(cfg, merged)
 	}
 }
 

internal/artifact/transcript_test.go

@@ -23,28 +23,28 @@ func TestFromMergedUsesBuildVersion(t *testing.T) {
 	}
 }
 
-func TestSelectedFromMergedDefaultsToSeriatimTranscript(t *testing.T) {
+func TestSelectedFromMergedDefaultsToIntermediateTranscript(t *testing.T) {
 	got := SelectedFromMerged(config.Config{}, model.MergedTranscript{})
+	if _, ok := got.(schema.IntermediateTranscript); !ok {
+		t.Fatalf("selected artifact type = %T, want schema.IntermediateTranscript", got)
+	}
+}
+
+func TestSelectedFromMergedUsesIntermediateWhenConfigured(t *testing.T) {
+	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaIntermediate}, model.MergedTranscript{})
+	if _, ok := got.(schema.IntermediateTranscript); !ok {
+		t.Fatalf("selected artifact type = %T, want schema.IntermediateTranscript", got)
+	}
+}
+
+func TestSelectedFromMergedUsesFullWhenConfigured(t *testing.T) {
+	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaFull}, model.MergedTranscript{})
 	if _, ok := got.(schema.Transcript); !ok {
 		t.Fatalf("selected artifact type = %T, want schema.Transcript", got)
 	}
 }
 
-func TestSelectedFromMergedUsesDefaultWhenConfigured(t *testing.T) {
+func TestIntermediateFromMergedEmitsOnlyIntermediateShape(t *testing.T) {
-	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaDefault}, model.MergedTranscript{})
-	if _, ok := got.(schema.DefaultTranscript); !ok {
-		t.Fatalf("selected artifact type = %T, want schema.DefaultTranscript", got)
-	}
-}
-
-func TestSelectedFromMergedUsesSeriatimWhenConfigured(t *testing.T) {
-	got := SelectedFromMerged(config.Config{OutputSchema: config.OutputSchemaSeriatim}, model.MergedTranscript{})
-	if _, ok := got.(schema.Transcript); !ok {
-		t.Fatalf("selected artifact type = %T, want schema.Transcript", got)
-	}
-}
-
-func TestDefaultFromMergedEmitsOnlyDefaultShape(t *testing.T) {
 	merged := model.MergedTranscript{
 		Segments: []model.Segment{
 			{
@@ -65,19 +65,19 @@ func TestDefaultFromMergedEmitsOnlyDefaultShape(t *testing.T) {
 		},
 	}
 
-	got := DefaultFromMerged(config.Config{OutputSchema: config.OutputSchemaDefault}, merged)
+	got := IntermediateFromMerged(config.Config{OutputSchema: config.OutputSchemaIntermediate}, merged)
-	want := schema.DefaultTranscript{
+	want := schema.IntermediateTranscript{
-		Metadata: schema.DefaultMetadata{
+		Metadata: schema.IntermediateMetadata{
 			Application:  ApplicationName,
 			Version:      buildinfo.Version,
-			OutputSchema: config.OutputSchemaDefault,
+			OutputSchema: config.OutputSchemaIntermediate,
 		},
-		Segments: []schema.DefaultSegment{
+		Segments: []schema.IntermediateSegment{
 			{ID: 1, Start: 1, End: 2, Speaker: "Alice", Text: "hello", Categories: []string{"backchannel"}},
 		},
 	}
 	if !reflect.DeepEqual(got, want) {
-		t.Fatalf("default transcript = %#v, want %#v", got, want)
+		t.Fatalf("intermediate transcript = %#v, want %#v", got, want)
 	}
 }
 

internal/builtin/postprocess.go

@@ -51,8 +51,8 @@ func (validateOutput) Process(ctx context.Context, in model.MergedTranscript, cf
 	selected := artifact.SelectedFromMerged(cfg, in)
 	var err error
 	switch transcript := selected.(type) {
-	case schema.DefaultTranscript:
+	case schema.IntermediateTranscript:
-		err = schema.ValidateDefaultTranscript(transcript)
+		err = schema.ValidateIntermediateTranscript(transcript)
 	case schema.MinimalTranscript:
 		err = schema.ValidateMinimalTranscript(transcript)
 	case schema.Transcript:

internal/builtin/postprocess_test.go

@@ -81,7 +81,7 @@ func TestValidateOutputUsesMinimalSchemaWhenConfigured(t *testing.T) {
 	}
 }
 
-func TestValidateOutputUsesSeriatimSchemaWhenConfigured(t *testing.T) {
+func TestValidateOutputUsesFullSchemaWhenConfigured(t *testing.T) {
 	merged := model.MergedTranscript{
 		Segments: []model.Segment{
 			{
@@ -100,7 +100,7 @@ func TestValidateOutputUsesSeriatimSchemaWhenConfigured(t *testing.T) {
 	}
 
 	cfg := testConfig()
-	cfg.OutputSchema = config.OutputSchemaSeriatim
+	cfg.OutputSchema = config.OutputSchemaFull
 	got, events, err := validateOutput{}.Process(context.Background(), merged, cfg)
 	if err != nil {
 		t.Fatalf("validate output: %v", err)

internal/cli/merge.go

@@ -15,7 +15,12 @@ func newMergeCommand() *cobra.Command {
 		Use:   "merge",
 		Short: "Run the transcript merge pipeline",
 		RunE: func(cmd *cobra.Command, args []string) error {
-			cfg, err := config.NewMergeConfig(opts)
+			mergeOpts := opts
+			if !cmd.Flags().Changed("output-schema") {
+				mergeOpts.OutputSchema = ""
+			}
+
+			cfg, err := config.NewMergeConfig(mergeOpts)
 			if err != nil {
 				return err
 			}
@@ -32,7 +37,7 @@ func newMergeCommand() *cobra.Command {
 	flags.StringVar(&opts.AutocorrectFile, "autocorrect", "", "autocorrect rules file")
 	flags.StringVar(&opts.InputReader, "input-reader", config.DefaultInputReader, "input reader module")
 	flags.StringVar(&opts.OutputModules, "output-modules", config.DefaultOutputModules, "comma-separated output modules")
-	flags.StringVar(&opts.OutputSchema, "output-schema", config.DefaultOutputSchema, "output JSON schema: default, minimal, or seriatim")
+	flags.StringVar(&opts.OutputSchema, "output-schema", config.DefaultOutputSchema, "output JSON schema: seriatim-minimal, seriatim-intermediate (default), or seriatim-full")
 	flags.StringVar(&opts.PreprocessingModules, "preprocessing-modules", config.DefaultPreprocessingModules, "comma-separated preprocessing modules")
 	flags.StringVar(&opts.PostprocessingModules, "postprocessing-modules", config.DefaultPostprocessingModules, "comma-separated postprocessing modules")
 	flags.StringVar(&opts.CoalesceGap, "coalesce-gap", config.DefaultCoalesceGapValue, "maximum same-speaker gap in seconds for coalesce")

internal/cli/merge_test.go

@@ -114,7 +114,8 @@ func TestMergeWritesMergedOutputAndReport(t *testing.T) {
 	}
 }
 
-func TestMergeWritesDefaultOutputSchema(t *testing.T) {
+func TestMergeWritesIntermediateOutputSchema(t *testing.T) {
+	t.Setenv(config.OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeJSONFile(t, dir, "input.json", `{
 		"segments": [
@@ -131,10 +132,10 @@ func TestMergeWritesDefaultOutputSchema(t *testing.T) {
 		t.Fatalf("merge failed: %v", err)
 	}
 
-	var transcript schema.DefaultTranscript
+	var transcript schema.IntermediateTranscript
 	readJSON(t, output, &transcript)
-	if transcript.Metadata.OutputSchema != config.OutputSchemaDefault {
+	if transcript.Metadata.OutputSchema != config.OutputSchemaIntermediate {
-		t.Fatalf("output_schema = %q, want default", transcript.Metadata.OutputSchema)
+		t.Fatalf("output_schema = %q, want %q", transcript.Metadata.OutputSchema, config.OutputSchemaIntermediate)
 	}
 	if len(transcript.Segments) != 1 {
 		t.Fatalf("segment count = %d, want 1", len(transcript.Segments))
@@ -149,12 +150,12 @@ func TestMergeWritesDefaultOutputSchema(t *testing.T) {
 	}
 	for _, forbidden := range []string{"overlap_groups", "source", "derived_from", "words"} {
 		if strings.Contains(string(outputBytes), forbidden) {
-			t.Fatalf("default output contains %q:\n%s", forbidden, outputBytes)
+			t.Fatalf("intermediate output contains %q:\n%s", forbidden, outputBytes)
 		}
 	}
 }
 
-func TestMergeWritesSeriatimOutputSchema(t *testing.T) {
+func TestMergeWritesFullOutputSchema(t *testing.T) {
 	dir := t.TempDir()
 	input := writeJSONFile(t, dir, "input.json", `{
 		"segments": [
@@ -166,7 +167,7 @@ func TestMergeWritesSeriatimOutputSchema(t *testing.T) {
 	err := executeMergeRaw(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "seriatim",
+		"--output-schema", config.OutputSchemaFull,
 	)
 	if err != nil {
 		t.Fatalf("merge failed: %v", err)
@@ -203,7 +204,7 @@ func TestMergeWritesMinimalOutputSchema(t *testing.T) {
 	err := executeMerge(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--report-file", reportPath,
 	)
 	if err != nil {
@@ -215,8 +216,8 @@ func TestMergeWritesMinimalOutputSchema(t *testing.T) {
 	if transcript.Metadata.Application != "seriatim" {
 		t.Fatalf("application = %q, want seriatim", transcript.Metadata.Application)
 	}
-	if transcript.Metadata.OutputSchema != "minimal" {
+	if transcript.Metadata.OutputSchema != config.OutputSchemaMinimal {
-		t.Fatalf("output_schema = %q, want minimal", transcript.Metadata.OutputSchema)
+		t.Fatalf("output_schema = %q, want %q", transcript.Metadata.OutputSchema, config.OutputSchemaMinimal)
 	}
 	if got, want := len(transcript.Segments), 2; got != want {
 		t.Fatalf("segment count = %d, want %d", got, want)
@@ -330,7 +331,7 @@ func TestMergeValidateMinimalOutputBeforeAssignIDsFails(t *testing.T) {
 	err := executeMerge(
 		"--input-file", input,
 		"--output-file", output,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--postprocessing-modules", "validate-output,assign-ids",
 	)
 	if err == nil {
@@ -472,16 +473,16 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 
 	var transcript model.FinalTranscript
 	readJSON(t, output, &transcript)
-	if len(transcript.OverlapGroups) != 0 {
+	if got, want := len(transcript.OverlapGroups), 1; got != want {
-		t.Fatalf("overlap groups = %#v, want none", transcript.OverlapGroups)
+		t.Fatalf("overlap group count = %d, want %d (%#v)", got, want, transcript.OverlapGroups)
 	}
-	if got, want := len(transcript.Segments), 3; got != want {
+	if got, want := len(transcript.Segments), 2; got != want {
 		t.Fatalf("segment count = %d, want %d", got, want)
 	}
 
-	wantTexts := []string{"outside hello there", "bob reply", "later"}
+	wantTexts := []string{"outside hello there later", "bob reply"}
-	wantSpeakers := []string{"Alice", "Bob", "Alice"}
+	wantSpeakers := []string{"Alice", "Bob"}
-	wantRefs := []string{"word-run:1:1:1", "word-run:1:2:1", "word-run:1:1:2"}
+	wantRefs := []string{"word-run:1:1:1", "word-run:1:2:1"}
 	for index, segment := range transcript.Segments {
 		if segment.ID != index+1 {
 			t.Fatalf("segment %d id = %d, want %d", index, segment.ID, index+1)
@@ -498,8 +499,8 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 		if segment.SourceSegmentIndex != nil {
 			t.Fatalf("segment %d source_segment_index = %d, want nil", index, *segment.SourceSegmentIndex)
 		}
-		if segment.OverlapGroupID != 0 {
+		if segment.OverlapGroupID != 1 {
-			t.Fatalf("segment %d overlap_group_id = %d, want 0", index, segment.OverlapGroupID)
+			t.Fatalf("segment %d overlap_group_id = %d, want 1", index, segment.OverlapGroupID)
 		}
 	}
 	if !equalStrings(transcript.Segments[0].DerivedFrom, []string{inputA + "#0"}) {
@@ -516,7 +517,7 @@ func TestMergeResolvesOverlapGroupsWithWordRuns(t *testing.T) {
 
 	var rpt report.Report
 	readJSON(t, reportPath, &rpt)
-	if !hasReportEvent(rpt, "postprocessing", "resolve-overlaps", "processed 1 overlap group(s); changed 1; removed 2 original segment(s); created 3 replacement segment(s)") {
+	if !hasReportEvent(rpt, "postprocessing", "resolve-overlaps", "processed 1 overlap group(s); changed 1; removed 2 original segment(s); created 2 replacement segment(s)") {
 		t.Fatal("expected resolve-overlaps summary report event")
 	}
 }
@@ -1927,7 +1928,7 @@ func TestMergeResolveOverlapsPreservesAbsorbedContextPrefix(t *testing.T) {
 		"--input-file", inputA,
 		"--input-file", inputB,
 		"--speakers", speakers,
-		"--output-schema", "minimal",
+		"--output-schema", config.OutputSchemaMinimal,
 		"--output-file", output,
 	)
 	if err != nil {
@@ -1994,7 +1995,7 @@ func executeMerge(args ...string) error {
 	if !hasOutputSchemaFlag(args) {
 		// Most integration tests were written against the full envelope; keep
 		// that behavior unless the caller explicitly asks for another schema.
-		args = append(args, "--output-schema", config.OutputSchemaSeriatim)
+		args = append(args, "--output-schema", config.OutputSchemaFull)
 	}
 	return executeMergeRaw(args...)
 }

internal/config/config.go

@@ -13,22 +13,23 @@ import (
 const (
 	DefaultInputReader            = "json-files"
 	DefaultOutputModules          = "json"
-	DefaultOutputSchema           = OutputSchemaDefault
+	DefaultOutputSchema           = OutputSchemaIntermediate
 	DefaultPreprocessingModules   = "validate-raw,normalize-speakers,trim-text"
 	DefaultPostprocessingModules  = "detect-overlaps,resolve-overlaps,backchannel,filler,resolve-danglers,coalesce,detect-overlaps,autocorrect,assign-ids,validate-output"
-	DefaultOverlapWordRunGap      = 0.75
+	DefaultOverlapWordRunGap      = 1.0
 	DefaultWordRunReorderWindow   = 1.0
 	DefaultCoalesceGap            = 3.0
 	DefaultCoalesceGapValue       = "3.0"
 	DefaultBackchannelMaxDuration = 2.0
 	DefaultFillerMaxDuration      = 1.25
+	OutputSchemaEnv               = "SERIATIM_OUTPUT_SCHEMA"
 	OverlapWordRunGapEnv          = "SERIATIM_OVERLAP_WORD_RUN_GAP"
 	WordRunReorderWindowEnv       = "SERIATIM_OVERLAP_WORD_RUN_REORDER_WINDOW"
 	BackchannelMaxDurationEnv     = "SERIATIM_BACKCHANNEL_MAX_DURATION"
 	FillerMaxDurationEnv          = "SERIATIM_FILLER_MAX_DURATION"
-	OutputSchemaDefault           = "default"
+	OutputSchemaMinimal           = "seriatim-minimal"
-	OutputSchemaSeriatim          = "seriatim"
+	OutputSchemaIntermediate      = "seriatim-intermediate"
-	OutputSchemaMinimal           = "minimal"
+	OutputSchemaFull              = "seriatim-full"
 )
 
 // MergeOptions captures raw CLI option values before validation.
@@ -70,7 +71,7 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
 	cfg := Config{
 		InputReader:            strings.TrimSpace(opts.InputReader),
 		OutputModules:          nil,
-		OutputSchema:           strings.TrimSpace(opts.OutputSchema),
+		OutputSchema:           "",
 		PreprocessingModules:   nil,
 		PostprocessingModules:  nil,
 		OverlapWordRunGap:      DefaultOverlapWordRunGap,
@@ -83,14 +84,12 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
 	if cfg.InputReader == "" {
 		return Config{}, errors.New("--input-reader is required")
 	}
-	if cfg.OutputSchema == "" {
+	var err error
-		cfg.OutputSchema = DefaultOutputSchema
+	cfg.OutputSchema, err = resolveOutputSchema(opts.OutputSchema)
-	}
+	if err != nil {
-	if err := validateOutputSchema(cfg.OutputSchema); err != nil {
 		return Config{}, err
 	}
 
-	var err error
 	cfg.OutputModules, err = parseModuleList(opts.OutputModules)
 	if err != nil {
 		return Config{}, fmt.Errorf("--output-modules: %w", err)
@@ -189,13 +188,27 @@ func parseModuleList(value string) ([]string, error) {
 
 func validateOutputSchema(value string) error {
 	switch value {
-	case OutputSchemaDefault, OutputSchemaSeriatim, OutputSchemaMinimal:
+	case OutputSchemaMinimal, OutputSchemaIntermediate, OutputSchemaFull:
 		return nil
 	default:
-		return fmt.Errorf("--output-schema must be one of %q, %q, or %q", OutputSchemaDefault, OutputSchemaMinimal, OutputSchemaSeriatim)
+		return fmt.Errorf("--output-schema must be one of %q, %q, or %q", OutputSchemaMinimal, OutputSchemaIntermediate, OutputSchemaFull)
 	}
 }
 
+func resolveOutputSchema(value string) (string, error) {
+	value = strings.TrimSpace(value)
+	if value == "" {
+		value = strings.TrimSpace(os.Getenv(OutputSchemaEnv))
+	}
+	if value == "" {
+		value = DefaultOutputSchema
+	}
+	if err := validateOutputSchema(value); err != nil {
+		return "", err
+	}
+	return value, nil
+}
+
 func normalizeInputFiles(paths []string) ([]string, error) {
 	if len(paths) == 0 {
 		return nil, errors.New("at least one --input-file is required")

internal/config/config_test.go

@@ -46,7 +46,8 @@ func TestDuplicateInputFilesFailValidation(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaDefaultsToDefault(t *testing.T) {
+func TestOutputSchemaDefaultsToIntermediate(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -67,7 +68,8 @@ func TestOutputSchemaDefaultsToDefault(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaAcceptsDefault(t *testing.T) {
+func TestOutputSchemaAcceptsIntermediate(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -77,19 +79,20 @@ func TestOutputSchemaAcceptsDefault(t *testing.T) {
 		OutputFile:            output,
 		InputReader:           DefaultInputReader,
 		OutputModules:         DefaultOutputModules,
-		OutputSchema:          OutputSchemaDefault,
+		OutputSchema:          OutputSchemaIntermediate,
 		PreprocessingModules:  DefaultPreprocessingModules,
 		PostprocessingModules: DefaultPostprocessingModules,
 	})
 	if err != nil {
 		t.Fatalf("config failed: %v", err)
 	}
-	if cfg.OutputSchema != OutputSchemaDefault {
+	if cfg.OutputSchema != OutputSchemaIntermediate {
-		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaDefault)
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaIntermediate)
 	}
 }
 
 func TestOutputSchemaAcceptsMinimal(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -111,7 +114,8 @@ func TestOutputSchemaAcceptsMinimal(t *testing.T) {
 	}
 }
 
-func TestOutputSchemaAcceptsSeriatim(t *testing.T) {
+func TestOutputSchemaAcceptsFull(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -121,19 +125,87 @@ func TestOutputSchemaAcceptsSeriatim(t *testing.T) {
 		OutputFile:            output,
 		InputReader:           DefaultInputReader,
 		OutputModules:         DefaultOutputModules,
-		OutputSchema:          OutputSchemaSeriatim,
+		OutputSchema:          OutputSchemaFull,
 		PreprocessingModules:  DefaultPreprocessingModules,
 		PostprocessingModules: DefaultPostprocessingModules,
 	})
 	if err != nil {
 		t.Fatalf("config failed: %v", err)
 	}
-	if cfg.OutputSchema != OutputSchemaSeriatim {
+	if cfg.OutputSchema != OutputSchemaFull {
-		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaSeriatim)
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaFull)
+	}
+}
+
+func TestOutputSchemaUsesEnvWhenFlagOmitted(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, OutputSchemaFull)
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	cfg, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err != nil {
+		t.Fatalf("config failed: %v", err)
+	}
+	if cfg.OutputSchema != OutputSchemaFull {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaFull)
+	}
+}
+
+func TestOutputSchemaFlagOverridesEnv(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, OutputSchemaFull)
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	cfg, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		OutputSchema:          OutputSchemaMinimal,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err != nil {
+		t.Fatalf("config failed: %v", err)
+	}
+	if cfg.OutputSchema != OutputSchemaMinimal {
+		t.Fatalf("output schema = %q, want %q", cfg.OutputSchema, OutputSchemaMinimal)
+	}
+}
+
+func TestOutputSchemaRejectsInvalidEnvValue(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "compact")
+	dir := t.TempDir()
+	input := writeTempFile(t, dir, "input.json")
+	output := filepath.Join(dir, "merged.json")
+
+	_, err := NewMergeConfig(MergeOptions{
+		InputFiles:            []string{input},
+		OutputFile:            output,
+		InputReader:           DefaultInputReader,
+		OutputModules:         DefaultOutputModules,
+		PreprocessingModules:  DefaultPreprocessingModules,
+		PostprocessingModules: DefaultPostprocessingModules,
+	})
+	if err == nil {
+		t.Fatal("expected output schema error")
+	}
+	if !strings.Contains(err.Error(), "--output-schema must be one of") {
+		t.Fatalf("unexpected error: %v", err)
 	}
 }
 
 func TestOutputSchemaRejectsUnknownValue(t *testing.T) {
+	t.Setenv(OutputSchemaEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")
 	output := filepath.Join(dir, "merged.json")
@@ -155,7 +227,36 @@ func TestOutputSchemaRejectsUnknownValue(t *testing.T) {
 	}
 }
 
-func TestOverlapWordRunGapDefaultsTo075(t *testing.T) {
+func TestOutputSchemaRejectsLegacyValues(t *testing.T) {
+	tests := []string{"default", "minimal", "seriatim"}
+
+	for _, legacy := range tests {
+		t.Run(legacy, func(t *testing.T) {
+			t.Setenv(OutputSchemaEnv, "")
+			dir := t.TempDir()
+			input := writeTempFile(t, dir, "input.json")
+			output := filepath.Join(dir, "merged.json")
+
+			_, err := NewMergeConfig(MergeOptions{
+				InputFiles:            []string{input},
+				OutputFile:            output,
+				InputReader:           DefaultInputReader,
+				OutputModules:         DefaultOutputModules,
+				OutputSchema:          legacy,
+				PreprocessingModules:  DefaultPreprocessingModules,
+				PostprocessingModules: DefaultPostprocessingModules,
+			})
+			if err == nil {
+				t.Fatal("expected output schema error")
+			}
+			if !strings.Contains(err.Error(), "--output-schema must be one of") {
+				t.Fatalf("unexpected error: %v", err)
+			}
+		})
+	}
+}
+
+func TestOverlapWordRunGapDefaultsTo1(t *testing.T) {
 	t.Setenv(OverlapWordRunGapEnv, "")
 	dir := t.TempDir()
 	input := writeTempFile(t, dir, "input.json")

internal/danglers/danglers.go

@@ -184,7 +184,7 @@ func nearestPriorMatch(segments []model.Segment, consumed []bool, index int) int
 		if consumed[candidate] {
 			continue
 		}
-		if sharesDerivedFrom(segments[index], segments[candidate]) {
+		if segments[index].Speaker == segments[candidate].Speaker && sharesDerivedFrom(segments[index], segments[candidate]) {
 			return candidate
 		}
 	}
@@ -196,7 +196,7 @@ func nearestSubsequentMatch(segments []model.Segment, consumed []bool, index int
 		if consumed[candidate] {
 			continue
 		}
-		if sharesDerivedFrom(segments[index], segments[candidate]) {
+		if segments[index].Speaker == segments[candidate].Speaker && sharesDerivedFrom(segments[index], segments[candidate]) {
 			return candidate
 		}
 	}

internal/danglers/danglers_test.go

@@ -51,6 +51,30 @@ func TestApplyUsesAnyDerivedFromIntersection(t *testing.T) {
 	assertSegment(t, got.Segments[0], "resolve-danglers:1", "target end.", 1, 4, []string{"source#1", "source#2", "source#3"})
 }
 
+func TestApplyDoesNotMergeDanglersAcrossSpeakersBackward(t *testing.T) {
+	in := transcript(
+		segment("a", "Alice", 1, 2, "target", []string{"source#1"}),
+		segment("b", "Bob", 3, 4, "end.", []string{"source#1"}),
+	)
+
+	got, summary := Apply(in)
+	if summary.DanglersMerged != 0 || !reflect.DeepEqual(got, in) {
+		t.Fatalf("unexpected merge:\ngot  %#v\nwant %#v", got, in)
+	}
+}
+
+func TestApplyDoesNotMergeDanglersAcrossSpeakersForward(t *testing.T) {
+	in := transcript(
+		segment("a", "Alice", 1, 2, "start", []string{"source#1"}),
+		segment("b", "Bob", 3, 4, "target", []string{"source#1"}),
+	)
+
+	got, summary := Apply(in)
+	if summary.DanglersMerged != 0 || !reflect.DeepEqual(got, in) {
+		t.Fatalf("unexpected merge:\ngot  %#v\nwant %#v", got, in)
+	}
+}
+
 func TestApplyDoesNotMergeWithoutSharedProvenance(t *testing.T) {
 	in := transcript(
 		segment("a", "Alice", 1, 2, "target", []string{"source#1"}),

schema/full-output.schema.json

@@ -1,6 +1,6 @@
 {
   "$schema": "https://json-schema.org/draft/2020-12/schema",
-  "$id": "https://gitea.maximumdirect.net/eric/seriatim/schema/output.schema.json",
+  "$id": "https://gitea.maximumdirect.net/eric/seriatim/schema/full-output.schema.json",
   "title": "seriatim full output transcript",
   "type": "object",
   "additionalProperties": false,

schema/intermediate-output.schema.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json-schema.org/draft/2020-12/schema",
-  "$id": "https://gitea.maximumdirect.net/eric/seriatim/schema/default-output.schema.json",
+  "$id": "https://gitea.maximumdirect.net/eric/seriatim/schema/intermediate-output.schema.json",
-  "title": "seriatim default output transcript",
+  "title": "seriatim intermediate output transcript",
   "type": "object",
   "additionalProperties": false,
   "required": ["metadata", "segments"],
@@ -13,7 +13,7 @@
       "properties": {
         "application": { "type": "string" },
         "version": { "type": "string" },
-        "output_schema": { "type": "string", "const": "default" }
+        "output_schema": { "type": "string", "const": "seriatim-intermediate" }
       }
     },
     "segments": {

schema/minimal-output.schema.json

@@ -13,7 +13,7 @@
       "properties": {
         "application": { "type": "string" },
         "version": { "type": "string" },
-        "output_schema": { "type": "string", "const": "minimal" }
+        "output_schema": { "type": "string", "const": "seriatim-minimal" }
       }
     },
     "segments": {

schema/output.go

@@ -14,9 +14,9 @@ import (
 var schemaFS embed.FS
 
 const (
-	outputSchemaPath        = "output.schema.json"
+	fullOutputSchemaPath         = "full-output.schema.json"
-	defaultOutputSchemaPath = "default-output.schema.json"
+	intermediateOutputSchemaPath = "intermediate-output.schema.json"
-	minimalOutputSchemaPath = "minimal-output.schema.json"
+	minimalOutputSchemaPath      = "minimal-output.schema.json"
 )
 
 var (
@@ -32,10 +32,10 @@ type Transcript struct {
 	OverlapGroups []OverlapGroup `json:"overlap_groups"`
 }
 
-// DefaultTranscript is seriatim's default public JSON output contract.
+// IntermediateTranscript is seriatim's intermediate public JSON output contract.
-type DefaultTranscript struct {
+type IntermediateTranscript struct {
-	Metadata DefaultMetadata  `json:"metadata"`
+	Metadata IntermediateMetadata  `json:"metadata"`
-	Segments []DefaultSegment `json:"segments"`
+	Segments []IntermediateSegment `json:"segments"`
 }
 
 // MinimalTranscript is seriatim's compact public JSON output contract.
@@ -55,8 +55,8 @@ type Metadata struct {
 	OutputModules         []string `json:"output_modules"`
 }
 
-// DefaultMetadata records default artifact identity.
+// IntermediateMetadata records intermediate artifact identity.
-type DefaultMetadata struct {
+type IntermediateMetadata struct {
 	Application  string `json:"application"`
 	Version      string `json:"version"`
 	OutputSchema string `json:"output_schema"`
@@ -84,9 +84,9 @@ type Segment struct {
 	OverlapGroupID     int      `json:"overlap_group_id,omitempty"`
 }
 
-// DefaultSegment is the compact public transcript segment shape with
+// IntermediateSegment is the compact public transcript segment shape with
 // categories.
-type DefaultSegment struct {
+type IntermediateSegment struct {
 	ID         int      `json:"id"`
 	Start      float64  `json:"start"`
 	End        float64  `json:"end"`
@@ -115,7 +115,7 @@ type OverlapGroup struct {
 	Resolution string   `json:"resolution"`
 }
 
-// ValidateTranscript validates a typed transcript against the public JSON
+// ValidateTranscript validates a full transcript against the public JSON
 // schema and seriatim-specific semantic rules.
 func ValidateTranscript(transcript Transcript) error {
 	if err := validateSemantics(transcript); err != nil {
@@ -129,18 +129,18 @@ func ValidateTranscript(transcript Transcript) error {
 	return ValidateJSON(data)
 }
 
-// ValidateDefaultTranscript validates the default transcript against the
+// ValidateIntermediateTranscript validates the intermediate transcript against
-// default JSON schema and seriatim-specific semantic rules.
+// the intermediate JSON schema and seriatim-specific semantic rules.
-func ValidateDefaultTranscript(transcript DefaultTranscript) error {
+func ValidateIntermediateTranscript(transcript IntermediateTranscript) error {
-	if err := validateDefaultSemantics(transcript); err != nil {
+	if err := validateIntermediateSemantics(transcript); err != nil {
 		return err
 	}
 
 	data, err := json.Marshal(transcript)
 	if err != nil {
-		return fmt.Errorf("marshal default transcript for schema validation: %w", err)
+		return fmt.Errorf("marshal intermediate transcript for schema validation: %w", err)
 	}
-	return ValidateDefaultJSON(data)
+	return ValidateIntermediateJSON(data)
 }
 
 // ValidateMinimalTranscript validates a minimal transcript against the minimal
@@ -159,13 +159,13 @@ func ValidateMinimalTranscript(transcript MinimalTranscript) error {
 
 // ValidateJSON validates serialized output JSON against the public schema.
 func ValidateJSON(data []byte) error {
-	return validateJSONWithSchema(data, outputSchemaPath)
+	return validateJSONWithSchema(data, fullOutputSchemaPath)
 }
 
-// ValidateDefaultJSON validates serialized default output JSON against the
+// ValidateIntermediateJSON validates serialized intermediate output JSON
-// default public schema.
+// against the intermediate public schema.
-func ValidateDefaultJSON(data []byte) error {
+func ValidateIntermediateJSON(data []byte) error {
-	return validateJSONWithSchema(data, defaultOutputSchemaPath)
+	return validateJSONWithSchema(data, intermediateOutputSchemaPath)
 }
 
 // ValidateMinimalJSON validates serialized minimal output JSON against the
@@ -245,7 +245,7 @@ func validateSemantics(transcript Transcript) error {
 	return nil
 }
 
-func validateDefaultSemantics(transcript DefaultTranscript) error {
+func validateIntermediateSemantics(transcript IntermediateTranscript) error {
 	for index, segment := range transcript.Segments {
 		wantID := index + 1
 		if segment.ID != wantID {

schema/output_test.go

@@ -21,11 +21,11 @@ func TestValidateMinimalTranscriptAcceptsValidTranscript(t *testing.T) {
 	}
 }
 
-func TestValidateDefaultTranscriptAcceptsValidTranscript(t *testing.T) {
+func TestValidateIntermediateTranscriptAcceptsValidTranscript(t *testing.T) {
-	transcript := validDefaultTranscript()
+	transcript := validIntermediateTranscript()
 
-	if err := ValidateDefaultTranscript(transcript); err != nil {
+	if err := ValidateIntermediateTranscript(transcript); err != nil {
-		t.Fatalf("validate default transcript: %v", err)
+		t.Fatalf("validate intermediate transcript: %v", err)
 	}
 }
 
@@ -34,7 +34,7 @@ func TestValidateMinimalJSONRejectsMissingRequiredField(t *testing.T) {
 		"metadata": {
 			"application": "seriatim",
 			"version": "dev",
-			"output_schema": "minimal"
+				"output_schema": "seriatim-minimal"
 		}
 	}`))
 	assertErrorContains(t, err, "segments")
@@ -45,7 +45,7 @@ func TestValidateMinimalJSONRejectsWrongFieldType(t *testing.T) {
 		"metadata": {
 			"application": "seriatim",
 			"version": "dev",
-			"output_schema": "minimal"
+				"output_schema": "seriatim-minimal"
 		},
 		"segments": [
 			{
@@ -68,7 +68,7 @@ func TestValidateMinimalJSONRejectsUnexpectedFields(t *testing.T) {
 		{
 			name: "top-level overlap groups",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "minimal"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-minimal"},
 				"segments": [],
 				"overlap_groups": []
 			}`,
@@ -76,28 +76,28 @@ func TestValidateMinimalJSONRejectsUnexpectedFields(t *testing.T) {
 		{
 			name: "segment source",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "minimal"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-minimal"},
 				"segments": [{"id": 1, "source": "input.json", "start": 1, "end": 2, "speaker": "Alice", "text": "hello"}]
 			}`,
 		},
 		{
 			name: "segment categories",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "minimal"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-minimal"},
 				"segments": [{"id": 1, "start": 1, "end": 2, "speaker": "Alice", "text": "hello", "categories": ["backchannel"]}]
 			}`,
 		},
 		{
 			name: "segment derived from",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "minimal"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-minimal"},
 				"segments": [{"id": 1, "start": 1, "end": 2, "speaker": "Alice", "text": "hello", "derived_from": ["input.json#0"]}]
 			}`,
 		},
 		{
 			name: "segment words",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "minimal"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-minimal"},
 				"segments": [{"id": 1, "start": 1, "end": 2, "speaker": "Alice", "text": "hello", "words": []}]
 			}`,
 		},
@@ -111,24 +111,24 @@ func TestValidateMinimalJSONRejectsUnexpectedFields(t *testing.T) {
 	}
 }
 
-func TestValidateDefaultJSONRejectsMissingRequiredField(t *testing.T) {
+func TestValidateIntermediateJSONRejectsMissingRequiredField(t *testing.T) {
-	err := ValidateDefaultJSON([]byte(`{
+	err := ValidateIntermediateJSON([]byte(`{
-		"metadata": {
+			"metadata": {
-			"application": "seriatim",
+				"application": "seriatim",
-			"version": "dev",
+				"version": "dev",
-			"output_schema": "default"
+				"output_schema": "seriatim-intermediate"
-		}
+			}
-	}`))
+		}`))
 	assertErrorContains(t, err, "segments")
 }
 
-func TestValidateDefaultJSONRejectsWrongFieldType(t *testing.T) {
+func TestValidateIntermediateJSONRejectsWrongFieldType(t *testing.T) {
-	err := ValidateDefaultJSON([]byte(`{
+	err := ValidateIntermediateJSON([]byte(`{
-		"metadata": {
+			"metadata": {
-			"application": "seriatim",
+				"application": "seriatim",
-			"version": "dev",
+				"version": "dev",
-			"output_schema": "default"
+				"output_schema": "seriatim-intermediate"
-		},
+			},
 		"segments": [
 			{
 				"id": "1",
@@ -142,7 +142,7 @@ func TestValidateDefaultJSONRejectsWrongFieldType(t *testing.T) {
 	assertErrorContains(t, err, "id")
 }
 
-func TestValidateDefaultJSONRejectsUnexpectedFields(t *testing.T) {
+func TestValidateIntermediateJSONRejectsUnexpectedFields(t *testing.T) {
 	tests := []struct {
 		name string
 		json string
@@ -150,7 +150,7 @@ func TestValidateDefaultJSONRejectsUnexpectedFields(t *testing.T) {
 		{
 			name: "top-level overlap groups",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "default"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-intermediate"},
 				"segments": [],
 				"overlap_groups": []
 			}`,
@@ -158,21 +158,21 @@ func TestValidateDefaultJSONRejectsUnexpectedFields(t *testing.T) {
 		{
 			name: "segment source",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "default"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-intermediate"},
 				"segments": [{"id": 1, "source": "input.json", "start": 1, "end": 2, "speaker": "Alice", "text": "hello"}]
 			}`,
 		},
 		{
 			name: "segment derived from",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "default"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-intermediate"},
 				"segments": [{"id": 1, "start": 1, "end": 2, "speaker": "Alice", "text": "hello", "derived_from": ["input.json#0"]}]
 			}`,
 		},
 		{
 			name: "segment words",
 			json: `{
-				"metadata": {"application": "seriatim", "version": "dev", "output_schema": "default"},
+					"metadata": {"application": "seriatim", "version": "dev", "output_schema": "seriatim-intermediate"},
 				"segments": [{"id": 1, "start": 1, "end": 2, "speaker": "Alice", "text": "hello", "words": []}]
 			}`,
 		},
@@ -180,7 +180,7 @@ func TestValidateDefaultJSONRejectsUnexpectedFields(t *testing.T) {
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
-			err := ValidateDefaultJSON([]byte(test.json))
+			err := ValidateIntermediateJSON([]byte(test.json))
 			assertErrorContains(t, err, "additional properties")
 		})
 	}
@@ -267,7 +267,7 @@ func TestValidateJSONRejectsUnexpectedInternalFields(t *testing.T) {
 	}
 }
 
-func TestValidateDefaultTranscriptRejectsMissingOrNonSequentialIDs(t *testing.T) {
+func TestValidateIntermediateTranscriptRejectsMissingOrNonSequentialIDs(t *testing.T) {
 	tests := []struct {
 		name string
 		ids  []int
@@ -280,10 +280,10 @@ func TestValidateDefaultTranscriptRejectsMissingOrNonSequentialIDs(t *testing.T)
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
-			transcript := validDefaultTranscript()
+			transcript := validIntermediateTranscript()
 			transcript.Segments = transcript.Segments[:0]
 			for index, id := range test.ids {
-				transcript.Segments = append(transcript.Segments, DefaultSegment{
+				transcript.Segments = append(transcript.Segments, IntermediateSegment{
 					ID:      id,
 					Start:   float64(index),
 					End:     float64(index) + 1,
@@ -292,18 +292,18 @@ func TestValidateDefaultTranscriptRejectsMissingOrNonSequentialIDs(t *testing.T)
 				})
 			}
 
-			err := ValidateDefaultTranscript(transcript)
+			err := ValidateIntermediateTranscript(transcript)
 			assertErrorContains(t, err, test.want)
 		})
 	}
 }
 
-func TestValidateDefaultTranscriptRejectsInvalidTiming(t *testing.T) {
+func TestValidateIntermediateTranscriptRejectsInvalidTiming(t *testing.T) {
-	transcript := validDefaultTranscript()
+	transcript := validIntermediateTranscript()
 	transcript.Segments[0].Start = 2
 	transcript.Segments[0].End = 1
 
-	err := ValidateDefaultTranscript(transcript)
+	err := ValidateIntermediateTranscript(transcript)
 	assertErrorContains(t, err, "segment 0 has end")
 }
 
@@ -411,7 +411,7 @@ func validMinimalTranscript() MinimalTranscript {
 		Metadata: MinimalMetadata{
 			Application:  "seriatim",
 			Version:      "dev",
-			OutputSchema: "minimal",
+			OutputSchema: "seriatim-minimal",
 		},
 		Segments: []MinimalSegment{
 			{
@@ -425,14 +425,14 @@ func validMinimalTranscript() MinimalTranscript {
 	}
 }
 
-func validDefaultTranscript() DefaultTranscript {
+func validIntermediateTranscript() IntermediateTranscript {
-	return DefaultTranscript{
+	return IntermediateTranscript{
-		Metadata: DefaultMetadata{
+		Metadata: IntermediateMetadata{
 			Application:  "seriatim",
 			Version:      "dev",
-			OutputSchema: "default",
+			OutputSchema: "seriatim-intermediate",
 		},
-		Segments: []DefaultSegment{
+		Segments: []IntermediateSegment{
 			{
 				ID:         1,
 				Start:      1,