Add Seriatim configuration contract
This commit is contained in:
@@ -52,9 +52,20 @@ type WhisperXConfig struct {
|
||||
|
||||
// SeriatimConfig configures seriatim adapter settings.
|
||||
type SeriatimConfig struct {
|
||||
BinaryPath string `yaml:"binary_path"`
|
||||
Timeout string `yaml:"timeout"`
|
||||
Args []string `yaml:"args"`
|
||||
Binary string `yaml:"binary"`
|
||||
Timeout string `yaml:"timeout"`
|
||||
OutputSchema string `yaml:"output_schema"`
|
||||
CoalesceGap *float64 `yaml:"coalesce_gap"`
|
||||
Report *bool `yaml:"report"`
|
||||
Env SeriatimEnvConfig `yaml:"env"`
|
||||
}
|
||||
|
||||
// SeriatimEnvConfig configures optional advanced seriatim environment tuning.
|
||||
type SeriatimEnvConfig struct {
|
||||
OverlapWordRunGap *float64 `yaml:"overlap_word_run_gap"`
|
||||
OverlapWordRunReorderWindow *float64 `yaml:"overlap_word_run_reorder_window"`
|
||||
BackchannelMaxDuration *float64 `yaml:"backchannel_max_duration"`
|
||||
FillerMaxDuration *float64 `yaml:"filler_max_duration"`
|
||||
}
|
||||
|
||||
// AuditaConfig configures audita adapter settings.
|
||||
|
||||
@@ -82,6 +82,7 @@ func applyPipelineDefaults(cfg *PipelineConfig) {
|
||||
return
|
||||
}
|
||||
applyWhisperXDefaults(&cfg.WhisperX)
|
||||
applySeriatimDefaults(&cfg.Seriatim)
|
||||
}
|
||||
|
||||
func applyWhisperXDefaults(cfg *WhisperXConfig) {
|
||||
@@ -109,3 +110,31 @@ func intPtr(v int) *int {
|
||||
p := v
|
||||
return &p
|
||||
}
|
||||
|
||||
func applySeriatimDefaults(cfg *SeriatimConfig) {
|
||||
if cfg == nil {
|
||||
return
|
||||
}
|
||||
if cfg.Timeout == "" {
|
||||
cfg.Timeout = "10m"
|
||||
}
|
||||
if cfg.OutputSchema == "" {
|
||||
cfg.OutputSchema = "seriatim-intermediate"
|
||||
}
|
||||
if cfg.CoalesceGap == nil {
|
||||
cfg.CoalesceGap = float64Ptr(3.0)
|
||||
}
|
||||
if cfg.Report == nil {
|
||||
cfg.Report = boolPtr(true)
|
||||
}
|
||||
}
|
||||
|
||||
func float64Ptr(v float64) *float64 {
|
||||
p := v
|
||||
return &p
|
||||
}
|
||||
|
||||
func boolPtr(v bool) *bool {
|
||||
p := v
|
||||
return &p
|
||||
}
|
||||
|
||||
@@ -23,7 +23,7 @@ func TestLoadAndValidate(t *testing.T) {
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
timeout: 30s
|
||||
binary: seriatim
|
||||
audita:
|
||||
timeout: 1h
|
||||
analyzer:
|
||||
@@ -93,6 +93,8 @@ unknown_field: true
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: ""
|
||||
inputs:
|
||||
@@ -108,6 +110,8 @@ inputs:
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -124,6 +128,8 @@ inputs:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: ://
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -141,6 +147,8 @@ inputs:
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
timeout: definitely-not-a-duration
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -158,6 +166,8 @@ inputs:
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
retry_delay: not-a-duration
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -175,6 +185,8 @@ inputs:
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
retries: -1
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -192,6 +204,8 @@ inputs:
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
concurrency: 0
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
@@ -202,6 +216,140 @@ inputs:
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.whisperx.concurrency must be > 0",
|
||||
},
|
||||
{
|
||||
name: "unknown seriatim field fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
bogus: true
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantLoadErr: "strict decode failed",
|
||||
},
|
||||
{
|
||||
name: "unknown seriatim env field fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
env:
|
||||
unknown_knob: 1.0
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantLoadErr: "strict decode failed",
|
||||
},
|
||||
{
|
||||
name: "missing seriatim binary fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
timeout: 10m
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.binary is required",
|
||||
},
|
||||
{
|
||||
name: "invalid seriatim timeout fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
timeout: not-a-duration
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.timeout must be a valid duration",
|
||||
},
|
||||
{
|
||||
name: "invalid seriatim output schema fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
output_schema: invalid-schema
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.output_schema must be one of: seriatim-minimal, seriatim-intermediate, seriatim-full",
|
||||
},
|
||||
{
|
||||
name: "negative seriatim coalesce gap fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
coalesce_gap: -0.1
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.coalesce_gap must be >= 0",
|
||||
},
|
||||
{
|
||||
name: "invalid seriatim env float fails",
|
||||
pipelineYAML: `workspace:
|
||||
root: /tmp/narratio
|
||||
whisperx:
|
||||
transcribe_url: https://transcription.ai.rakestrawhome.com/transcribe
|
||||
seriatim:
|
||||
binary: seriatim
|
||||
env:
|
||||
overlap_word_run_gap: 0
|
||||
`,
|
||||
sessionYAML: `session_id: 2026-05-03
|
||||
inputs:
|
||||
audio_dir: ./audio
|
||||
speakers_file: ./speakers.yml
|
||||
autocorrect_file: ./autocorrect.yml
|
||||
glossary_file: ./glossary.yml
|
||||
`,
|
||||
wantValidate: "pipeline config \"pipeline.yml\" invalid: pipeline.seriatim.env.overlap_word_run_gap must be > 0 when provided",
|
||||
},
|
||||
}
|
||||
|
||||
for _, tt := range tests {
|
||||
@@ -247,6 +395,18 @@ inputs:
|
||||
if cfg.Pipeline.WhisperX.Concurrency == nil || *cfg.Pipeline.WhisperX.Concurrency != 2 {
|
||||
t.Fatalf("whisperx.concurrency = %v, want 2", cfg.Pipeline.WhisperX.Concurrency)
|
||||
}
|
||||
if cfg.Pipeline.Seriatim.Timeout != "10m" {
|
||||
t.Fatalf("seriatim.timeout = %q, want %q", cfg.Pipeline.Seriatim.Timeout, "10m")
|
||||
}
|
||||
if cfg.Pipeline.Seriatim.OutputSchema != "seriatim-intermediate" {
|
||||
t.Fatalf("seriatim.output_schema = %q, want %q", cfg.Pipeline.Seriatim.OutputSchema, "seriatim-intermediate")
|
||||
}
|
||||
if cfg.Pipeline.Seriatim.CoalesceGap == nil || *cfg.Pipeline.Seriatim.CoalesceGap != 3.0 {
|
||||
t.Fatalf("seriatim.coalesce_gap = %v, want 3.0", cfg.Pipeline.Seriatim.CoalesceGap)
|
||||
}
|
||||
if cfg.Pipeline.Seriatim.Report == nil || *cfg.Pipeline.Seriatim.Report != true {
|
||||
t.Fatalf("seriatim.report = %v, want true", cfg.Pipeline.Seriatim.Report)
|
||||
}
|
||||
}
|
||||
|
||||
err = Validate(cfg)
|
||||
@@ -280,6 +440,13 @@ func TestValidateMissingAudioSource(t *testing.T) {
|
||||
RetryDelay: "2s",
|
||||
Concurrency: &concurrency,
|
||||
},
|
||||
Seriatim: SeriatimConfig{
|
||||
Binary: "seriatim",
|
||||
Timeout: "10m",
|
||||
OutputSchema: "seriatim-intermediate",
|
||||
CoalesceGap: float64Ptr(3.0),
|
||||
Report: boolPtr(true),
|
||||
},
|
||||
},
|
||||
Session: &SessionConfig{
|
||||
SessionID: "2026-05-03",
|
||||
|
||||
@@ -36,8 +36,7 @@ func validatePipeline(cfg *PipelineConfig) error {
|
||||
if err := validateWhisperX(cfg.WhisperX); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
if err := validateDuration("pipeline.seriatim.timeout", cfg.Seriatim.Timeout); err != nil {
|
||||
if err := validateSeriatim(cfg.Seriatim); err != nil {
|
||||
return err
|
||||
}
|
||||
if err := validateDuration("pipeline.audita.timeout", cfg.Audita.Timeout); err != nil {
|
||||
@@ -85,6 +84,44 @@ func validateWhisperX(cfg WhisperXConfig) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
func validateSeriatim(cfg SeriatimConfig) error {
|
||||
if strings.TrimSpace(cfg.Binary) == "" {
|
||||
return fmt.Errorf("pipeline.seriatim.binary is required")
|
||||
}
|
||||
if err := validateDuration("pipeline.seriatim.timeout", cfg.Timeout); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
switch cfg.OutputSchema {
|
||||
case "seriatim-minimal", "seriatim-intermediate", "seriatim-full":
|
||||
default:
|
||||
return fmt.Errorf("pipeline.seriatim.output_schema must be one of: seriatim-minimal, seriatim-intermediate, seriatim-full")
|
||||
}
|
||||
|
||||
if cfg.CoalesceGap == nil {
|
||||
return fmt.Errorf("pipeline.seriatim.coalesce_gap must be set (defaults should populate this)")
|
||||
}
|
||||
if *cfg.CoalesceGap < 0 {
|
||||
return fmt.Errorf("pipeline.seriatim.coalesce_gap must be >= 0")
|
||||
}
|
||||
|
||||
for _, item := range []struct {
|
||||
name string
|
||||
value *float64
|
||||
}{
|
||||
{name: "pipeline.seriatim.env.overlap_word_run_gap", value: cfg.Env.OverlapWordRunGap},
|
||||
{name: "pipeline.seriatim.env.overlap_word_run_reorder_window", value: cfg.Env.OverlapWordRunReorderWindow},
|
||||
{name: "pipeline.seriatim.env.backchannel_max_duration", value: cfg.Env.BackchannelMaxDuration},
|
||||
{name: "pipeline.seriatim.env.filler_max_duration", value: cfg.Env.FillerMaxDuration},
|
||||
} {
|
||||
if item.value != nil && *item.value <= 0 {
|
||||
return fmt.Errorf("%s must be > 0 when provided", item.name)
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func validateSession(cfg *SessionConfig) error {
|
||||
if strings.TrimSpace(cfg.SessionID) == "" {
|
||||
return fmt.Errorf("session.session_id is required")
|
||||
|
||||
Reference in New Issue
Block a user