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
import { apiJson, apiFetch, apiPost } from './client';
import type { SystemInfo, ModelStatus, LogsResponse, ClearTauriResponse } from './types';
// ── Tauri IPC helpers ────────────────────────────────────────────────────
// Try native Tauri invoke() first — it's faster (no HTTP round-trip) and
// works when the Python backend is still booting. Falls back to HTTP when
// running in browser dev mode (no Tauri shell).
// ── System info (polled every 5s) ────────────────────────────────────────
export interface SysinfoData {
cpu: number;
ram: number;
total_ram: number;
vram: number;
gpu_active: boolean;
}
// Cache VRAM from Python — it changes much slower than CPU/RAM, so we
// only refresh it every 15s instead of every 5s poll cycle.
let _vramCache: { vram: number; gpu_active: boolean; ts: number } | null = null;
const VRAM_CACHE_TTL = 15_000;
export async function sysinfo(): Promise<SysinfoData> {
// Rust provides CPU + RAM; VRAM stays at 0. We merge with the Python
// endpoint to get GPU data when available.
const rustData = await apiJson<SysinfoData>('/sysinfo');
// If we got data from Rust (vram=0), enrich with Python's VRAM data
// but only re-fetch every 15s to avoid hammering the backend.
if (rustData.vram === 0) {
const now = Date.now();
if (!_vramCache || now - _vramCache.ts > VRAM_CACHE_TTL) {
try {
const pyData = await apiJson<SysinfoData>('/sysinfo');
_vramCache = { vram: pyData.vram, gpu_active: pyData.gpu_active, ts: now };
} catch {
// Python backend not ready yet — return Rust-only data
return rustData;
}
}
return {
...rustData,
vram: _vramCache.vram,
gpu_active: _vramCache.gpu_active,
};
}
return rustData;
}
// ── Model status ─────────────────────────────────────────────────────────
export async function modelStatus(): Promise<ModelStatus> {
return apiJson<ModelStatus>('/model/status');
}
// ── Loaded-model residency (MM2-04 endpoints) ────────────────────────────
/** One entry from GET /model/loaded — a model currently resident in memory.
* `engine_id`/`is_active_engine` attribute TTS-family entries to an engine
* (a model can stay resident after the user switches engines). */
export interface LoadedModel {
id: string; // 'tts' | 'asr' | 'diarization' | 'sidecar:<e>' | 'engine:<e>' | 'capture-asr'
name: string;
checkpoint: string;
device: string;
vram_mb: number;
unloadable: boolean;
note?: string;
engine_id?: string;
is_active_engine?: boolean | null;
}
/** Free/total memory snapshot from GET /model/loaded. RAM is always present;
* VRAM fields appear only on a dedicated-GPU host; `warning` is a low-memory
* advisory string when free memory is below the headroom threshold. */
export interface SystemMemory {
ram_available_gb?: number;
ram_total_gb?: number;
vram_free_gb?: number;
vram_total_gb?: number;
warning?: string;
}
export interface LoadedModelsResponse {
models: LoadedModel[];
count: number;
system?: SystemMemory;
}
export async function listLoadedModels(): Promise<LoadedModelsResponse> {
return apiJson<LoadedModelsResponse>('/model/loaded');
}
/** Unload one resident model by its /model/loaded `id`. The model reloads
* lazily on next use — unloading only frees memory, it never loses data. */
export async function unloadLoadedModel(modelId: string): Promise<unknown> {
return apiPost(`/model/unload/${encodeURIComponent(modelId)}`);
}
// ── Audio cleaning ───────────────────────────────────────────────────────
export async function cleanAudio(formData: FormData): Promise<Response> {
// Returns Response because caller needs blob body + X-Clean-Filename header.
return apiFetch('/clean-audio', { method: 'POST', body: formData });
}
// ── System info (one-shot, for Settings) ─────────────────────────────────
export async function systemInfo(): Promise<SystemInfo> {
return apiJson<SystemInfo>('/system/info');
}
// ── Notifications (polled by header bell + logs footer) ──────────────────
interface SystemNotification {
id: string;
level: 'info' | 'warn' | 'error';
title?: string;
message?: string;
action?: { type: string; target: string; label?: string };
}
export interface NotificationsResponse {
notifications: SystemNotification[];
}
export async function systemNotifications(): Promise<NotificationsResponse> {
return apiJson<NotificationsResponse>('/system/notifications');
}
// ── Logs (polled every 5s) ───────────────────────────────────────────────
export async function systemLogs(tail: number = 300): Promise<LogsResponse> {
return apiJson<LogsResponse>(`/system/logs?tail=${tail}`);
}
export async function systemLogsTauri(tail: number = 300): Promise<LogsResponse> {
return apiJson<LogsResponse>(`/system/logs/tauri?tail=${tail}`);
}
// ── Log clearing ─────────────────────────────────────────────────────────
export async function clearSystemLogs(): Promise<unknown> {
return apiPost('/system/logs/clear');
}
export async function clearTauriLogs(): Promise<ClearTauriResponse> {
return apiPost<ClearTauriResponse>('/system/logs/tauri/clear');
}
// ── Memory flush ─────────────────────────────────────────────────────────
export async function flushMemory(unloadModel: boolean = false): Promise<unknown> {
return apiPost(`/system/flush-memory?unload_model=${unloadModel}`);
}