Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ Local-first speech-to-text CLI.
- single-instance command coordination via unix socket
- audio capture via PipeWire/Pulse
- streaming ASR via NVIDIA Riva gRPC
- transcript normalization + optional trailing space
- transcript normalization + sentence capitalization + optional trailing space
- output adapters:
- clipboard command (`clipboard_cmd`)
- optional paste command override (`paste_cmd`)
Expand Down
5 changes: 4 additions & 1 deletion apps/sotto/internal/config/defaults.go
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,10 @@ func Default() Config {
LanguageCode: "en-US",
Model: "",
},
Transcript: TranscriptConfig{TrailingSpace: true},
Transcript: TranscriptConfig{
TrailingSpace: true,
CapitalizeSentences: true,
},
Indicator: IndicatorConfig{
Enable: true,
Backend: "hypr",
Expand Down
12 changes: 9 additions & 3 deletions apps/sotto/internal/config/parser_jsonc.go
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,8 @@ type jsoncASR struct {
}

type jsoncTranscript struct {
TrailingSpace *bool `json:"trailing_space"`
TrailingSpace *bool `json:"trailing_space"`
CapitalizeSentences *bool `json:"capitalize_sentences"`
}

type jsoncIndicator struct {
Expand Down Expand Up @@ -176,8 +177,13 @@ func (payload jsoncConfig) applyTo(cfg *Config) ([]Warning, error) {
}
}

if payload.Transcript != nil && payload.Transcript.TrailingSpace != nil {
cfg.Transcript.TrailingSpace = *payload.Transcript.TrailingSpace
if payload.Transcript != nil {
if payload.Transcript.TrailingSpace != nil {
cfg.Transcript.TrailingSpace = *payload.Transcript.TrailingSpace
}
if payload.Transcript.CapitalizeSentences != nil {
cfg.Transcript.CapitalizeSentences = *payload.Transcript.CapitalizeSentences
}
}

if payload.Indicator != nil {
Expand Down
6 changes: 6 additions & 0 deletions apps/sotto/internal/config/parser_legacy.go
Original file line number Diff line number Diff line change
Expand Up @@ -218,6 +218,12 @@ func applyRootKey(cfg *Config, key, value string) error {
return fmt.Errorf("invalid bool for transcript.trailing_space: %w", err)
}
cfg.Transcript.TrailingSpace = b
case "transcript.capitalize_sentences":
b, err := strconv.ParseBool(value)
if err != nil {
return fmt.Errorf("invalid bool for transcript.capitalize_sentences: %w", err)
}
cfg.Transcript.CapitalizeSentences = b
case "indicator.enable":
b, err := strconv.ParseBool(value)
if err != nil {
Expand Down
12 changes: 12 additions & 0 deletions apps/sotto/internal/config/parser_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -175,6 +175,18 @@ func TestParsePasteShortcut(t *testing.T) {
}
}

func TestParseTranscriptCapitalizeSentencesJSONC(t *testing.T) {
cfg, _, err := Parse(`{"transcript":{"capitalize_sentences":false}}`, Default())
require.NoError(t, err)
require.False(t, cfg.Transcript.CapitalizeSentences)
}

func TestParseTranscriptCapitalizeSentencesLegacy(t *testing.T) {
cfg, _, err := Parse("transcript.capitalize_sentences = false\n", Default())
require.NoError(t, err)
require.False(t, cfg.Transcript.CapitalizeSentences)
}

func TestParseIndicatorBackend(t *testing.T) {
cfg, _, err := Parse(`
{
Expand Down
3 changes: 2 additions & 1 deletion apps/sotto/internal/config/types.go
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,8 @@ type ASRConfig struct {

// TranscriptConfig controls transcript assembly formatting.
type TranscriptConfig struct {
TrailingSpace bool
TrailingSpace bool
CapitalizeSentences bool
}

// IndicatorConfig controls visual indicator and audio cue behavior.
Expand Down
5 changes: 4 additions & 1 deletion apps/sotto/internal/pipeline/transcriber.go
Original file line number Diff line number Diff line change
Expand Up @@ -189,7 +189,10 @@ func (t *Transcriber) StopAndTranscribe(ctx context.Context) (session.StopResult
return result, fmt.Errorf("collect final transcript: %w", err)
}

transcribed := transcript.Assemble(segments, t.cfg.Transcript.TrailingSpace)
transcribed := transcript.Assemble(segments, transcript.Options{
TrailingSpace: t.cfg.Transcript.TrailingSpace,
CapitalizeSentences: t.cfg.Transcript.CapitalizeSentences,
})
rawPCM := capture.RawPCM()
t.writeDebugAudio(rawPCM)
t.closeDebugArtifacts()
Expand Down
2 changes: 1 addition & 1 deletion apps/sotto/internal/pipeline/transcriber_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -260,7 +260,7 @@ func TestStopAndTranscribeSuccessPath(t *testing.T) {

result, err := transcriber.StopAndTranscribe(context.Background())
require.NoError(t, err)
require.Equal(t, "hello world ", result.Transcript)
require.Equal(t, "Hello world ", result.Transcript)
require.Equal(t, "Mic (mic-1)", result.AudioDevice)
require.Equal(t, int64(4096), result.BytesCaptured)
require.Equal(t, 12*time.Millisecond, result.GRPCLatency)
Expand Down
24 changes: 21 additions & 3 deletions apps/sotto/internal/transcript/assemble.go
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,14 @@ package transcript

import "strings"

// Assemble joins final ASR segments and applies whitespace/trailing-space normalization.
func Assemble(finalSegments []string, trailingSpace bool) string {
// Options controls transcript assembly formatting behavior.
type Options struct {
TrailingSpace bool
CapitalizeSentences bool
}

// Assemble joins final ASR segments and applies configured normalization.
func Assemble(finalSegments []string, opts Options) string {
if len(finalSegments) == 0 {
return ""
}
Expand All @@ -15,8 +21,20 @@ func Assemble(finalSegments []string, trailingSpace bool) string {
return ""
}

if trailingSpace {
if opts.CapitalizeSentences {
normalized = capitalizeSentences(normalized)
}

if opts.TrailingSpace {
return normalized + " "
}
return normalized
}

func capitalizeSentences(text string) string {
text = capitalizeSentenceStarts(text)
text = pronounIContractionPattern.ReplaceAllStringFunc(text, func(match string) string {
return "I" + match[1:]
})
return capitalizeStandalonePronounI(text)
}
Loading
Loading