VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
Python aiaudiobookcudadubbingelevenlabs-alternativehuggingfacelocal-firstmlxomnivoice-studiospeech-to-texttext-to-speechtranscriptiontranslatettsvoice-aivoice-cloningvoice-generationvoicestudioworkflow
/**
* Store wiring for the backend-backed dictation prefs (prefsSlice).
*
* Asserts the contract the Voice panel + CaptureWidget depend on:
* • loadDictationPrefs() hydrates enabled/mode/model from GET /dictation/prefs
* • each setter is optimistic AND write-throughs to POST /dictation/prefs
* • the store re-syncs from the POST response (so a server-side normalisation
* like repo_id → canonical id can't leave the UI out of step)
* • a failed write rolls the optimistic value back
*/
import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
// Mock the api client BEFORE importing the store (prefsSlice imports it at
// module load). Each test swaps the implementations via the spies below.
const apiJson = vi.fn();
const apiPost = vi.fn();
vi.mock('../api/client', () => ({
apiJson: (...a: any[]) => apiJson(...a),
apiPost: (...a: any[]) => apiPost(...a),
}));
import { useAppStore } from '../store';
import { createPrefsSlice } from '../store/prefsSlice';
function flush() {
// Let the write-through promise (.then) settle.
return new Promise((r) => setTimeout(r, 0));
}
describe('dictation prefs store wiring', () => {
it('seeds the cross-platform default before backend hydration', () => {
const slice = createPrefsSlice(vi.fn() as any, vi.fn() as any, {} as any);
expect(slice.dictationModelId).toBe('sherpa-whisper-tiny');
});
beforeEach(() => {
apiJson.mockReset();
apiPost.mockReset();
// Reset to slice seeds.
useAppStore.setState({
dictationEnabled: true,
dictationMode: 'toggle',
dictationModelId: 'sherpa-whisper-tiny',
dictationLoaded: false,
});
});
afterEach(() => vi.restoreAllMocks());
it('hydrates from GET /dictation/prefs', async () => {
apiJson.mockResolvedValue({ enabled: false, mode: 'hold', model_id: 'sherpa-whisper-tiny' });
await useAppStore.getState().loadDictationPrefs();
const s = useAppStore.getState();
expect(apiJson).toHaveBeenCalledWith('/dictation/prefs');
expect(s.dictationEnabled).toBe(false);
expect(s.dictationMode).toBe('hold');
expect(s.dictationModelId).toBe('sherpa-whisper-tiny');
expect(s.dictationLoaded).toBe(true);
});
it('still marks loaded when the backend route fails (older build)', async () => {
apiJson.mockRejectedValue(new Error('404'));
await useAppStore.getState().loadDictationPrefs();
expect(useAppStore.getState().dictationLoaded).toBe(true);
// Seeds preserved.
expect(useAppStore.getState().dictationMode).toBe('toggle');
});
it('setDictationMode is optimistic and write-throughs, then re-syncs', async () => {
apiPost.mockResolvedValue({ enabled: true, mode: 'hold', model_id: 'sherpa-parakeet-tdt-v3' });
useAppStore.getState().setDictationMode('hold');
// Optimistic update happened synchronously.
expect(useAppStore.getState().dictationMode).toBe('hold');
await flush();
expect(apiPost).toHaveBeenCalledWith('/dictation/prefs', { mode: 'hold' });
expect(useAppStore.getState().dictationMode).toBe('hold');
});
it('re-syncs the model id from the POST response (repo_id → canonical id)', async () => {
// The user picks via repo_id; the backend normalises to the canonical id.
apiPost.mockResolvedValue({
enabled: true,
mode: 'toggle',
model_id: 'sherpa-parakeet-tdt-v2',
});
useAppStore
.getState()
.setDictationModelId('csukuangfj/sherpa-onnx-nemo-parakeet-tdt-0.6b-v2-int8');
await flush();
expect(useAppStore.getState().dictationModelId).toBe('sherpa-parakeet-tdt-v2');
});
it('rolls back the optimistic value when the write fails', async () => {
apiPost.mockRejectedValue(new Error('boom'));
useAppStore.getState().setDictationEnabled(false);
expect(useAppStore.getState().dictationEnabled).toBe(false); // optimistic
await flush();
expect(useAppStore.getState().dictationEnabled).toBe(true); // rolled back
});
});