Skip to content

Build your personal knowledge base with Trilium Notes

feat(transcription): turn speaker identification off in the settings

Elian Doran531cf6374f87
diff --git a/.claude/skills/working-with-transcription/SKILL.md b/.claude/skills/working-with-transcription/SKILL.md
index c5738c77bc..5cdfe09876 100644
--- a/.claude/skills/working-with-transcription/SKILL.md
+++ b/.claude/skills/working-with-transcription/SKILL.md
@@ -55,7 +55,7 @@ Nothing the transcription process imports may reach `@triliumnext/core` or `@tri
## The model
-Users choose the recognizer under Options → Transcription: the non-synced option `transcriptionModel` holds a key of `TRANSCRIPTION_MODELS` (commons `transcript.ts`), `parakeet-int8` (the default, "Standard") or `parakeet-fp32` ("High accuracy"), each naming its resource and precision; an unknown value reads as the default (`getTranscriptionModel()`). The launcher calls `requireResources(getTranscriptionResources(<option>))` before it creates the meeting note, which shows the download dialog for those missing, and `startRecording()` resolves the same option through `getModelPaths()`. The models are resources of the resource manager (`apps/server/src/services/resources.ts`), each in `<data dir>/resources/<id>-<hash>/` and checked against the SHA-256 in `model.ts`: `transcription.parakeet` (int8, 670 MB), `transcription.parakeet-fp32` (2.5 GB, pinned to a Hugging Face revision), `transcription.silero-vad` (0.6 MB) and `transcription.titanet` (40 MB). The hash in the directory name covers every file's path and SHA-256, so changing a file of one model downloads that model alone again, into a new directory. A recording needs Parakeet and Silero; without TitaNet it labels no speakers.
+Users choose the recognizer under Options → Transcription: the non-synced option `transcriptionModel` holds a key of `TRANSCRIPTION_MODELS` (commons `transcript.ts`), `parakeet-int8` (the default, "Standard") or `parakeet-fp32` ("High accuracy"), each naming its resource and precision; an unknown value reads as the default (`getTranscriptionModel()`). The synced option `transcriptionSpeakersEnabled` (the "Identify speakers" switch, default on) decides whether a recording labels speakers. The launcher calls `requireResources(getTranscriptionResources(<model>, <speakers>))` before it creates the meeting note, which shows the download dialog for those missing (TitaNet only while the switch is on), and `startRecording()` resolves the same two options through `getModelPaths()`, which passes `speaker: null` while the switch is off. The models are resources of the resource manager (`apps/server/src/services/resources.ts`), each in `<data dir>/resources/<id>-<hash>/` and checked against the SHA-256 in `model.ts`: `transcription.parakeet` (int8, 670 MB), `transcription.parakeet-fp32` (2.5 GB, pinned to a Hugging Face revision), `transcription.silero-vad` (0.6 MB) and `transcription.titanet` (40 MB). The hash in the directory name covers every file's path and SHA-256, so changing a file of one model downloads that model alone again, into a new directory. A recording needs Parakeet and Silero; without TitaNet it labels no speakers.
A directory installed by hand, `<data dir>/transcription-model` (`LOCAL_MODEL_DIR`), takes precedence over the downloads, each model for itself, when it holds that model's files (each Parakeet precision when every one of its files is there, Silero when `silero_vad.onnx` is, TitaNet when `speaker/nemo_en_titanet_small.onnx` is); for `pnpm desktop:start` that is `apps/desktop/data/transcription-model`, which the scripts below read. `ModelPaths.recognizer` carries the precision with the directory, and `loadSherpaEngine()` loads that precision's files; it records it as `precision` in the engine settings and applies `blankPenalty` to int8 only (see "Findings so far").
diff --git a/apps/client/src/translations/en/translation.json b/apps/client/src/translations/en/translation.json
index bc90595c6d..978f662346 100644
--- a/apps/client/src/translations/en/translation.json
+++ b/apps/client/src/translations/en/translation.json
@@ -4629,6 +4629,8 @@
"european_languages": "European languages",
"european_languages_description": "Parakeet TDT 0.6B v3 by NVIDIA recognizes 25 European languages and works out which one is spoken. High accuracy makes fewer mistakes, but it is a 2.4 GB download instead of 0.6 GB and needs about 1.6 GB more memory while recording. A model that is not downloaded yet is downloaded when you start a recording.",
"model_standard": "Standard",
- "model_high_accuracy": "High accuracy"
+ "model_high_accuracy": "High accuracy",
+ "identify_speakers": "Identify speakers",
+ "identify_speakers_description": "Tells the other participants apart and labels each of their utterances with its speaker; what you say into your microphone is always yours. Turn it off if it mixes up voices, or to save the processing it takes during a recording. It applies from the next recording."
}
}
diff --git a/apps/client/src/widgets/launch_bar/RecordMeetingButton.spec.tsx b/apps/client/src/widgets/launch_bar/RecordMeetingButton.spec.tsx
index 0ea8060998..693b2b9575 100644
--- a/apps/client/src/widgets/launch_bar/RecordMeetingButton.spec.tsx
+++ b/apps/client/src/widgets/launch_bar/RecordMeetingButton.spec.tsx
@@ -10,7 +10,8 @@ const mocks = vi.hoisted(() => ({
createNote: vi.fn(),
showError: vi.fn(),
requireResources: vi.fn(),
- transcriptionModel: null as string | null
+ transcriptionModel: null as string | null,
+ speakersEnabled: true
}));
vi.mock("../../services/date_notes", () => ({
@@ -23,7 +24,10 @@ vi.mock("../../services/dialog", () => ({
requireResources: mocks.requireResources
}));
vi.mock("../../services/options", () => ({
- default: { get: (name: string) => name === "transcriptionModel" ? mocks.transcriptionModel : null }
+ default: {
+ get: (name: string) => name === "transcriptionModel" ? mocks.transcriptionModel : null,
+ is: (name: string) => name === "transcriptionSpeakersEnabled" && mocks.speakersEnabled
+ }
}));
vi.mock("../../services/toast", () => ({
default: { showError: mocks.showError }
@@ -43,6 +47,7 @@ import RecordMeetingButton, { createMeetingNote, toggleRecording } from "./Recor
beforeEach(() => {
vi.clearAllMocks();
mocks.transcriptionModel = null;
+ mocks.speakersEnabled = true;
vi.useFakeTimers({ now: new Date(2026, 9, 13, 14, 30), toFake: [ "Date" ] });
});
@@ -122,6 +127,13 @@ describe("toggleRecording", () => {
[ "transcription.parakeet-fp32", "transcription.silero-vad", "transcription.titanet" ]);
});
+ it("leaves the speaker model out while speaker identification is off", async () => {
+ mocks.speakersEnabled = false;
+ await toggleRecording("idle");
+ expect(mocks.requireResources).toHaveBeenCalledExactlyOnceWith(
+ [ "transcription.parakeet", "transcription.silero-vad" ]);
+ });
+
it("starts nothing when no meeting note could be created", async () => {
mocks.getInboxNote.mockResolvedValue(null);
vi.spyOn(console, "warn").mockImplementation(() => {});
diff --git a/apps/client/src/widgets/launch_bar/RecordMeetingButton.tsx b/apps/client/src/widgets/launch_bar/RecordMeetingButton.tsx
index 0d4d435739..83866bfc80 100644
--- a/apps/client/src/widgets/launch_bar/RecordMeetingButton.tsx
+++ b/apps/client/src/widgets/launch_bar/RecordMeetingButton.tsx
@@ -105,7 +105,9 @@ export async function toggleRecording(status: TranscriptionStatus) {
if (status !== "idle") {
return;
}
- if (!await requireResources(getTranscriptionResources(options.get("transcriptionModel")))) {
+ const resources = getTranscriptionResources(
+ options.get("transcriptionModel"), options.is("transcriptionSpeakersEnabled"));
+ if (!await requireResources(resources)) {
return;
}
diff --git a/apps/client/src/widgets/type_widgets/options/transcription.spec.tsx b/apps/client/src/widgets/type_widgets/options/transcription.spec.tsx
index 769bfd43ba..869c2fbe2c 100644
--- a/apps/client/src/widgets/type_widgets/options/transcription.spec.tsx
+++ b/apps/client/src/widgets/type_widgets/options/transcription.spec.tsx
@@ -5,7 +5,9 @@ import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
const mocks = vi.hoisted(() => ({
electron: true,
stored: "" as string,
- saved: [] as string[]
+ saved: [] as string[],
+ speakersEnabled: true,
+ savedSpeakers: [] as boolean[]
}));
vi.mock("../../../services/utils", async (importOriginal) => ({
@@ -20,6 +22,10 @@ vi.mock("../../react/hooks", async (importOriginal) => ({
useTriliumOption: (name: string) => [
name === "transcriptionModel" ? mocks.stored : "",
(value: string) => void mocks.saved.push(value)
+ ],
+ useTriliumOptionBool: (name: string) => [
+ name === "transcriptionSpeakersEnabled" && mocks.speakersEnabled,
+ (value: boolean) => void mocks.savedSpeakers.push(value)
]
}));
@@ -33,6 +39,8 @@ beforeEach(() => {
mocks.electron = true;
mocks.stored = "parakeet-fp32";
mocks.saved = [];
+ mocks.speakersEnabled = true;
+ mocks.savedSpeakers = [];
host = document.body.appendChild(document.createElement("div"));
});
@@ -67,6 +75,21 @@ describe("TranscriptionSettings", () => {
expect(mocks.saved).toEqual([ "parakeet-int8" ]);
});
+ it("turns speaker identification off", () => {
+ open();
+ const toggle = host.querySelector<HTMLInputElement>("input.switch-toggle[id^='identify-speakers-']");
+ if (!toggle) {
+ throw new Error("The speaker switch is missing.");
+ }
+ expect(toggle.checked).toBe(true);
+
+ act(() => {
+ toggle.checked = false;
+ toggle.dispatchEvent(new Event("input", { bubbles: true }));
+ });
+ expect(mocks.savedSpeakers).toEqual([ false ]);
+ });
+
it("shows the default model for a value it does not know", () => {
mocks.stored = "whisper";
expect(open()?.value).toBe("parakeet-int8");
diff --git a/apps/client/src/widgets/type_widgets/options/transcription.tsx b/apps/client/src/widgets/type_widgets/options/transcription.tsx
index b78fa0f5e5..d25243e3b2 100644
--- a/apps/client/src/widgets/type_widgets/options/transcription.tsx
+++ b/apps/client/src/widgets/type_widgets/options/transcription.tsx
@@ -4,7 +4,8 @@ import { t } from "../../../services/i18n";
import { isElectron } from "../../../services/utils";
import { Card, OptionCardSection } from "../../react/Card";
import FormSelect from "../../react/FormSelect";
-import { useTriliumOption } from "../../react/hooks";
+import FormToggle from "../../react/FormToggle";
+import { useTriliumOption, useTriliumOptionBool } from "../../react/hooks";
import NoItems from "../../react/NoItems";
import OptionsPageHeader from "./components/OptionsPageHeader";
@@ -28,6 +29,7 @@ export default function TranscriptionSettings() {
function RecognitionSettings() {
const [ model, setModel ] = useTriliumOption("transcriptionModel");
+ const [ speakersEnabled, setSpeakersEnabled ] = useTriliumOptionBool("transcriptionSpeakersEnabled");
const labels: Record<TranscriptionModelId, string> = {
"parakeet-int8": t("transcription_options.model_standard"),
"parakeet-fp32": t("transcription_options.model_high_accuracy")
@@ -49,6 +51,14 @@ function RecognitionSettings() {
values={Object.values(TRANSCRIPTION_MODELS).map(({ id }) => ({ id, label: labels[id] }))}
/>
</OptionCardSection>
+
+ <OptionCardSection
+ name="identify-speakers"
+ label={t("transcription_options.identify_speakers")}
+ description={t("transcription_options.identify_speakers_description")}
+ >
+ <FormToggle currentValue={speakersEnabled} onChange={setSpeakersEnabled} />
+ </OptionCardSection>
</Card>
);
}
diff --git a/apps/desktop/src/services/transcription/index.spec.ts b/apps/desktop/src/services/transcription/index.spec.ts
index 7ad6223b15..2c9ca98d39 100644
--- a/apps/desktop/src/services/transcription/index.spec.ts
+++ b/apps/desktop/src/services/transcription/index.spec.ts
@@ -17,6 +17,7 @@ const h = vi.hoisted(() => ({
isDev: false,
installedResources: [] as string[],
transcriptionModel: null as string | null,
+ speakersEnabled: true,
registerResource: vi.fn(),
log: { info: vi.fn(), error: vi.fn() },
listeners: null as WorkerListeners | null,
@@ -63,7 +64,8 @@ vi.mock("@triliumnext/core", () => ({
},
getLog: () => h.log,
options: {
- getOptionOrNull: (name: string) => name === "transcriptionModel" ? h.transcriptionModel : null
+ getOptionOrNull: (name: string) => name === "transcriptionModel" ? h.transcriptionModel : null,
+ getOptionBool: (name: string) => name === "transcriptionSpeakersEnabled" && h.speakersEnabled
},
protected_session: { isProtectedSessionAvailable: () => h.isProtectedSessionAvailable },
utils: { isDev: () => h.isDev }
@@ -134,6 +136,7 @@ beforeEach(() => {
h.broadcast.length = 0;
h.isDev = false;
h.transcriptionModel = null;
+ h.speakersEnabled = true;
h.installedResources = [
"transcription.parakeet", "transcription.silero-vad", "transcription.titanet"
];
@@ -353,6 +356,13 @@ describe("transcription IPC", () => {
expect(h.worker.start.mock.calls[0][0].models.speaker).toBeNull();
});
+ it("records without the speaker model while speaker identification is off", async () => {
+ const invoke = await loadService();
+ h.speakersEnabled = false;
+ expect(await invoke("transcription-start", "meeting")).toBe("");
+ expect(h.worker.start.mock.calls[0][0].models.speaker).toBeNull();
+ });
+
it("records with the model the option selects, refusing while it is not downloaded", async () => {
const invoke = await loadService();
h.transcriptionModel = "parakeet-fp32";
diff --git a/apps/desktop/src/services/transcription/index.ts b/apps/desktop/src/services/transcription/index.ts
index 8afa740192..5b04aa81ed 100644
--- a/apps/desktop/src/services/transcription/index.ts
+++ b/apps/desktop/src/services/transcription/index.ts
@@ -62,7 +62,10 @@ export async function startRecording(noteId: unknown): Promise<string> {
if (typeof noteId !== "string" || !becca.getNote(noteId)) {
return t("transcription.note-not-found");
}
- const models = getModelPaths(options.getOptionOrNull("transcriptionModel"));
+ const models = getModelPaths(
+ options.getOptionOrNull("transcriptionModel"),
+ options.getOptionBool("transcriptionSpeakersEnabled")
+ );
if (!models) {
return t("transcription.model-not-downloaded");
}
diff --git a/apps/desktop/src/services/transcription/model.ts b/apps/desktop/src/services/transcription/model.ts
index 74160c944f..c01506d4af 100644
--- a/apps/desktop/src/services/transcription/model.ts
+++ b/apps/desktop/src/services/transcription/model.ts
@@ -131,13 +131,14 @@ export const TRANSCRIPTION_MODELS = [ ...RECOGNIZER_MODELS, VAD_MODEL, SPEAKER_M
/**
* Returns where the installed models are for a recording with the model `modelId` (a key of
* `TRANSCRIPTION_MODELS` in commons), or `null` while its recognizer or the VAD is missing.
- * Without the speaker model, the transcript labels no speakers.
+ * The speaker model is left out unless `identifySpeakers` holds and it is installed; without it,
+ * the transcript labels no speakers.
*/
-export function getModelPaths(modelId: string | null): ModelPaths | null {
+export function getModelPaths(modelId: string | null, identifySpeakers: boolean): ModelPaths | null {
const model = getTranscriptionModel(modelId);
const recognizer = resources.getPath(model.resourceId);
const vad = resources.getPath(VAD_MODEL.id);
- const speaker = resources.getPath(SPEAKER_MODEL.id);
+ const speaker = identifySpeakers ? resources.getPath(SPEAKER_MODEL.id) : null;
if (!recognizer || !vad) {
return null;
}
diff --git a/packages/commons/src/lib/options_interface.ts b/packages/commons/src/lib/options_interface.ts
index aa533a61d6..1cd07e700e 100644
--- a/packages/commons/src/lib/options_interface.ts
+++ b/packages/commons/src/lib/options_interface.ts
@@ -209,6 +209,8 @@ export interface OptionDefinitions extends KeyboardShortcutsOptions<KeyboardActi
* each device downloads and runs its own.
*/
transcriptionModel: string;
+ /** Whether a recording tells the other participants apart, labeling each utterance with its speaker. */
+ transcriptionSpeakersEnabled: boolean;
leftPaneVisible: boolean;
rightPaneVisible: boolean;
nativeTitleBarVisible: boolean;
diff --git a/packages/commons/src/lib/transcript.spec.ts b/packages/commons/src/lib/transcript.spec.ts
index 58f99e2a66..b18cffbe8f 100644
--- a/packages/commons/src/lib/transcript.spec.ts
+++ b/packages/commons/src/lib/transcript.spec.ts
@@ -95,10 +95,12 @@ describe("transcription models", () => {
expect(getTranscriptionModel("toString").id).toBe("parakeet-int8");
});
- it("needs the selected recognizer, the VAD and the speaker model to record", () => {
- expect(getTranscriptionResources("parakeet-fp32"))
+ it("needs the selected recognizer, the VAD and, to identify speakers, the speaker model", () => {
+ expect(getTranscriptionResources("parakeet-fp32", true))
.toEqual([ "transcription.parakeet-fp32", "transcription.silero-vad", "transcription.titanet" ]);
- expect(getTranscriptionResources(undefined))
+ expect(getTranscriptionResources(undefined, true))
.toEqual([ "transcription.parakeet", "transcription.silero-vad", "transcription.titanet" ]);
+ expect(getTranscriptionResources("parakeet-int8", false))
+ .toEqual([ "transcription.parakeet", "transcription.silero-vad" ]);
});
});
diff --git a/packages/commons/src/lib/transcript.ts b/packages/commons/src/lib/transcript.ts
index 60db8b3e1a..d924e680ba 100644
--- a/packages/commons/src/lib/transcript.ts
+++ b/packages/commons/src/lib/transcript.ts
@@ -74,9 +74,16 @@ export function getTranscriptionModel(id: string | null | undefined): Transcript
: TRANSCRIPTION_MODELS[DEFAULT_TRANSCRIPTION_MODEL];
}
-/** Returns the resources a recording with the model `modelId` needs. */
-export function getTranscriptionResources(modelId: string | null | undefined) {
- return [ getTranscriptionModel(modelId).resourceId, ...Object.values(TRANSCRIPTION_RESOURCES) ];
+/**
+ * Returns the resources a recording with the model `modelId` needs: the speaker model only when it
+ * identifies speakers.
+ */
+export function getTranscriptionResources(modelId: string | null | undefined, identifySpeakers: boolean) {
+ return [
+ getTranscriptionModel(modelId).resourceId,
+ TRANSCRIPTION_RESOURCES.vad,
+ ...(identifySpeakers ? [ TRANSCRIPTION_RESOURCES.speaker ] : [])
+ ];
}
/** Reads a transcript attachment's content, or returns `null` for content of another shape. */
diff --git a/packages/trilium-core/src/routes/api/options.ts b/packages/trilium-core/src/routes/api/options.ts
index 93ca6adaeb..99a19b3aab 100644
--- a/packages/trilium-core/src/routes/api/options.ts
+++ b/packages/trilium-core/src/routes/api/options.ts
@@ -65,6 +65,7 @@ const ALLOWED_OPTIONS = new Set<OptionNames>([
"codeLineWrapEnabled",
"transcriptCleanupEnabled",
"transcriptionModel",
+ "transcriptionSpeakersEnabled",
"codeNotesMimeTypes",
"contentManagerSortOrder",
"contentManagerViewMode",
diff --git a/packages/trilium-core/src/services/options_init.ts b/packages/trilium-core/src/services/options_init.ts
index ffc9ed8830..0bcaebce7b 100644
--- a/packages/trilium-core/src/services/options_init.ts
+++ b/packages/trilium-core/src/services/options_init.ts
@@ -185,6 +185,7 @@ const defaultOptions: DefaultOption[] = [
{ name: "codeLineWrapEnabled", value: "true", isSynced: false },
{ name: "transcriptCleanupEnabled", value: "true", isSynced: true },
{ name: "transcriptionModel", value: DEFAULT_TRANSCRIPTION_MODEL, isSynced: false },
+ { name: "transcriptionSpeakersEnabled", value: "true", isSynced: true },
{ name: "codeNoteTabWidth", value: "4", isSynced: true },
{ name: "codeNoteIndentWithTabs", value: "false", isSynced: true },
{