Added a module to coalesce adjacent same-speaker segments

This commit is contained in:
2026-04-27 19:30:00 -05:00
parent 13d972cb24
commit aab6d12730
12 changed files with 919 additions and 28 deletions

View File

@@ -14,9 +14,13 @@ const (
DefaultInputReader = "json-files"
DefaultOutputModules = "json"
DefaultPreprocessingModules = "validate-raw,normalize-speakers,trim-text"
DefaultPostprocessingModules = "detect-overlaps,resolve-overlaps,detect-overlaps,autocorrect,assign-ids,validate-output"
DefaultPostprocessingModules = "detect-overlaps,resolve-overlaps,coalesce,detect-overlaps,autocorrect,assign-ids,validate-output"
DefaultOverlapWordRunGap = 0.75
DefaultWordRunReorderWindow = 0.4
DefaultCoalesceGap = 3.0
DefaultCoalesceGapValue = "3.0"
OverlapWordRunGapEnv = "SERIATIM_OVERLAP_WORD_RUN_GAP"
WordRunReorderWindowEnv = "SERIATIM_OVERLAP_WORD_RUN_REORDER_WINDOW"
)
// MergeOptions captures raw CLI option values before validation.
@@ -30,6 +34,7 @@ type MergeOptions struct {
OutputModules string
PreprocessingModules string
PostprocessingModules string
CoalesceGap string
}
// Config is the validated runtime configuration for a merge invocation.
@@ -44,6 +49,8 @@ type Config struct {
PreprocessingModules []string
PostprocessingModules []string
OverlapWordRunGap float64
WordRunReorderWindow float64
CoalesceGap float64
}
// NewMergeConfig validates raw merge options and returns normalized config.
@@ -54,6 +61,8 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
PreprocessingModules: nil,
PostprocessingModules: nil,
OverlapWordRunGap: DefaultOverlapWordRunGap,
WordRunReorderWindow: DefaultWordRunReorderWindow,
CoalesceGap: DefaultCoalesceGap,
}
if cfg.InputReader == "" {
@@ -119,6 +128,14 @@ func NewMergeConfig(opts MergeOptions) (Config, error) {
if err != nil {
return Config{}, err
}
cfg.WordRunReorderWindow, err = parseWordRunReorderWindow()
if err != nil {
return Config{}, err
}
cfg.CoalesceGap, err = parseCoalesceGap(opts.CoalesceGap)
if err != nil {
return Config{}, err
}
return cfg, nil
}
@@ -198,21 +215,45 @@ func requireFile(path string, flag string) error {
}
func parseOverlapWordRunGap() (float64, error) {
value := strings.TrimSpace(os.Getenv(OverlapWordRunGapEnv))
return parsePositiveFloatEnv(OverlapWordRunGapEnv, DefaultOverlapWordRunGap)
}
func parseWordRunReorderWindow() (float64, error) {
return parsePositiveFloatEnv(WordRunReorderWindowEnv, DefaultWordRunReorderWindow)
}
func parseCoalesceGap(value string) (float64, error) {
value = strings.TrimSpace(value)
if value == "" {
return DefaultOverlapWordRunGap, nil
return DefaultCoalesceGap, nil
}
gap, err := strconv.ParseFloat(value, 64)
if err != nil {
return 0, fmt.Errorf("%s must be a positive number of seconds: %w", OverlapWordRunGapEnv, err)
return 0, fmt.Errorf("--coalesce-gap must be a non-negative number of seconds: %w", err)
}
if gap <= 0 {
return 0, fmt.Errorf("%s must be positive", OverlapWordRunGapEnv)
if gap < 0 {
return 0, fmt.Errorf("--coalesce-gap must be non-negative")
}
return gap, nil
}
func parsePositiveFloatEnv(name string, defaultValue float64) (float64, error) {
value := strings.TrimSpace(os.Getenv(name))
if value == "" {
return defaultValue, nil
}
parsed, err := strconv.ParseFloat(value, 64)
if err != nil {
return 0, fmt.Errorf("%s must be a positive number of seconds: %w", name, err)
}
if parsed <= 0 {
return 0, fmt.Errorf("%s must be positive", name)
}
return parsed, nil
}
func contains(values []string, target string) bool {
for _, value := range values {
if value == target {