@nebutra/voice-realtime 0.1.2 → 0.1.4

package/CHANGELOG.md

@@ -1,5 +1,15 @@
 # @nebutra/voice-realtime
 
+## 0.1.4
+
+### Patch Changes
+
+- Updated dependencies []:
+  - @nebutra/capability-kit@2.0.0
+  - @nebutra/errors@2.0.0
+  - @nebutra/audio-pipeline@0.1.4
+  - @nebutra/generation-context@0.1.4
+
 ## 0.1.2
 
 ### Patch Changes

package/package.json

@@ -1,11 +1,12 @@
 {
   "name": "@nebutra/voice-realtime",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "description": "Voice session lifecycle, narration synthesis, and enrollment surface",
   "private": false,
   "license": "MIT",
   "type": "module",
   "nebutra": {
+    "graph": "labs",
     "status": "wip",
     "productionReady": false,
     "surface": "generation-capability",
@@ -23,16 +24,20 @@
     "category": "ai",
     "summary": "Realtime voice session and narration synthesis surface"
   },
-  "main": "./src/index.ts",
-  "types": "./src/index.ts",
+  "main": "./dist/index.js",
+  "types": "./dist/index.d.ts",
   "exports": {
-    ".": "./src/index.ts"
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js",
+      "default": "./dist/index.js"
+    }
   },
   "dependencies": {
-    "@nebutra/audio-pipeline": "0.1.2",
-    "@nebutra/errors": "0.1.2",
-    "@nebutra/generation-context": "0.1.2",
-    "@nebutra/capability-kit": "0.2.2"
+    "@nebutra/audio-pipeline": "0.1.4",
+    "@nebutra/capability-kit": "2.0.0",
+    "@nebutra/errors": "2.0.0",
+    "@nebutra/generation-context": "0.1.4"
   },
   "devDependencies": {
     "@types/node": "^25.9.1",
@@ -53,6 +58,12 @@
   "publishConfig": {
     "access": "public"
   },
+  "files": [
+    "dist",
+    "README.md",
+    "LICENSE",
+    "CHANGELOG.md"
+  ],
   "scripts": {
     "build": "tsup",
     "test": "vitest run",

package/.nebutra/debug/voice-realtime.jsonl

@@ -1,5 +0,0 @@
-{"at":"2026-05-18T04:49:12.879Z","type":"session","session":{"id":"voice_session_thread_1_mpaq5ewf","tenantId":"tenant_a","threadId":"thread_1","room":"voice_tenant_a_thread_1","state":"listening"}}
-{"at":"2026-05-18T04:52:55.989Z","type":"session","session":{"id":"voice_session_thread_1_mpaqa71w","tenantId":"tenant_a","threadId":"thread_1","room":"voice_tenant_a_thread_1","state":"listening"}}
-{"at":"2026-05-18T05:20:20.680Z","type":"session","session":{"id":"voice_session_thread_1_mpar9g3s","tenantId":"tenant_a","threadId":"thread_1","room":"voice_tenant_a_thread_1","state":"listening"}}
-{"at":"2026-05-30T21:06:23.065Z","type":"session","session":{"id":"voice_session_thread_1_mpsuca8p","tenantId":"tenant_a","threadId":"thread_1","room":"voice_tenant_a_thread_1","state":"listening"}}
-{"at":"2026-07-01T01:38:46.615Z","type":"session","session":{"id":"voice_session_thread_1_mr1epzns","tenantId":"tenant_a","threadId":"thread_1","room":"voice_tenant_a_thread_1","state":"listening"}}

package/.turbo/turbo-build.log

@@ -1,22 +0,0 @@
-
-> @nebutra/voice-realtime@0.1.2 build /Users/tseka_luk/Documents/Nebutra-SaaS-Lab/Nebutra-Sailor/packages/ai/voice-realtime
-> tsup
-
-CLI Building entry: src/cli.ts, src/index.ts
-CLI Using tsconfig: tsconfig.json
-CLI tsup v8.5.1
-CLI Using tsup config: /Users/tseka_luk/Documents/Nebutra-SaaS-Lab/Nebutra-Sailor/packages/ai/voice-realtime/tsup.config.ts
-CLI Target: es2022
-CLI Cleaning output folder
-ESM Build start
-ESM dist/cli.js                1.32 KB
-ESM dist/index.js              152.00 B
-ESM dist/chunk-GARLTI7J.js     3.28 KB
-ESM dist/cli.js.map            2.34 KB
-ESM dist/index.js.map          71.00 B
-ESM dist/chunk-GARLTI7J.js.map 6.94 KB
-ESM ⚡️ Build success in 38ms
-DTS Build start
-DTS ⚡️ Build success in 2771ms
-DTS dist/cli.d.ts   13.00 B
-DTS dist/index.d.ts 1.74 KB

package/.turbo/turbo-test.log

@@ -1,14 +0,0 @@
-
-> @nebutra/voice-realtime@0.1.1 test /Users/tseka_luk/Documents/Nebutra-SaaS-Lab/Nebutra-Sailor/packages/ai/voice-realtime
-> vitest run
-
-
- RUN  v4.1.4 /Users/tseka_luk/Documents/Nebutra-SaaS-Lab/Nebutra-Sailor/packages/ai/voice-realtime
-
- ✓ src/index.test.ts (2 tests) 69ms
-
- Test Files  1 passed (1)
-      Tests  2 passed (2)
-   Start at  09:38:45
-   Duration  1.14s (transform 146ms, setup 0ms, import 184ms, tests 69ms, environment 0ms)
-

package/.turbo/turbo-typecheck.log

@@ -1,4 +0,0 @@
-
-> @nebutra/voice-realtime@0.1.2 typecheck /Users/tseka_luk/Documents/Nebutra-SaaS-Lab/Nebutra-Sailor/packages/ai/voice-realtime
-> tsc --noEmit
-

package/examples/enroll.ts

@@ -1,5 +0,0 @@
-import { VoiceRealtime } from "../src/index";
-
-const profile = await new VoiceRealtime().enroll({ tenantId: "demo", samplePaths: [] });
-
-process.stdout.write(`${profile.id}\n`);

package/examples/narration.ts

@@ -1,9 +0,0 @@
-import { createDemoBrandContext } from "@nebutra/generation-context";
-import { VoiceRealtime } from "../src/index";
-
-const asset = await new VoiceRealtime().synthesizeNarration(
-  { script: "Loop makes debugging visible.", targetDurationS: 3 },
-  createDemoBrandContext(),
-);
-
-process.stdout.write(`${asset.path}\n`);

package/examples/session.ts

@@ -1,5 +0,0 @@
-import { VoiceRealtime } from "../src/index";
-
-process.stdout.write(
-  `${JSON.stringify(await new VoiceRealtime().startSession({ tenantId: "demo", threadId: "thread_1" }), null, 2)}\n`,
-);

package/src/cli.ts

@@ -1,32 +0,0 @@
-import { createDemoBrandContext } from "@nebutra/generation-context";
-import { readVoiceDebug, VoiceRealtime } from "./index";
-
-const command = process.argv[2] ?? "doctor";
-const voice = new VoiceRealtime();
-
-if (command === "doctor") {
-  process.stdout.write(
-    `${JSON.stringify({ capability: "voice-realtime", results: await voice.doctor() }, null, 2)}\n`,
-  );
-} else if (command === "debug") {
-  process.stdout.write(
-    `${JSON.stringify({ capability: "voice-realtime", entries: await readVoiceDebug() }, null, 2)}\n`,
-  );
-} else if (command === "enroll") {
-  process.stdout.write(
-    `${JSON.stringify({ capability: "voice-realtime", profile: await voice.enroll({ tenantId: "local" }) }, null, 2)}\n`,
-  );
-} else if (command === "test-mic") {
-  process.stdout.write(
-    `${JSON.stringify({ capability: "voice-realtime", mic: await voice.testMic() }, null, 2)}\n`,
-  );
-} else if (command === "narrate") {
-  const asset = await voice.synthesizeNarration(
-    { script: "Loop helps indie developers debug the work that matters.", targetDurationS: 3 },
-    createDemoBrandContext(),
-  );
-  process.stdout.write(`${JSON.stringify({ capability: "voice-realtime", asset }, null, 2)}\n`);
-} else {
-  process.stderr.write(`Unknown voice-realtime command: ${command}\n`);
-  process.exitCode = 1;
-}

package/src/index.test.ts

@@ -1,33 +0,0 @@
-import { mkdtemp, readFile, rm } from "node:fs/promises";
-import { tmpdir } from "node:os";
-import { join } from "node:path";
-import { createDemoBrandContext } from "@nebutra/generation-context";
-import { afterEach, describe, expect, it } from "vitest";
-import { readVoiceDebug, VoiceRealtime } from "./index";
-
-let root: string | undefined;
-
-afterEach(async () => {
-  if (root) await rm(root, { recursive: true, force: true });
-  root = undefined;
-});
-
-describe("VoiceRealtime", () => {
-  it("starts a thread-bound voice session without importing the runtime", async () => {
-    const session = await new VoiceRealtime().startSession({
-      tenantId: "tenant_a",
-      threadId: "thread_1",
-    });
-    expect(session.state).toBe("listening");
-  });
-
-  it("synthesizes narration with BrandContext", async () => {
-    root = await mkdtemp(join(tmpdir(), "voice-realtime-"));
-    const asset = await new VoiceRealtime({ root }).synthesizeNarration(
-      { script: "Ship the story with the founder voice.", targetDurationS: 2 },
-      createDemoBrandContext(),
-    );
-    expect((await readFile(asset.path)).subarray(0, 4).toString()).toBe("RIFF");
-    expect(await readVoiceDebug(root)).toHaveLength(1);
-  });
-});

package/src/index.ts

@@ -1,158 +0,0 @@
-import { mkdir, writeFile } from "node:fs/promises";
-import { dirname, join } from "node:path";
-import { createToneWav } from "@nebutra/audio-pipeline";
-import { appendCapabilityDebug, readCapabilityDebug } from "@nebutra/capability-kit/debug";
-import {
-  assetId,
-  type BrandContext,
-  type GeneratedAsset,
-  requireBrandContext,
-} from "@nebutra/generation-context";
-
-export type VoiceState = "listening" | "thinking" | "speaking" | "paused" | "closed";
-
-export interface VoiceSession {
-  readonly id: string;
-  readonly tenantId: string;
-  readonly threadId: string;
-  readonly room: string;
-  readonly state: VoiceState;
-}
-
-export interface NarrationRequest {
-  readonly script: string;
-  readonly voiceProfileId?: string;
-  readonly targetDurationS?: number;
-}
-
-export interface VoiceAsset extends GeneratedAsset {
-  readonly kind: "voice";
-  readonly transcript: string;
-  readonly durationS: number;
-  readonly format: "wav";
-  readonly voiceProfileId: string;
-}
-
-export interface VoiceProfile {
-  readonly id: string;
-  readonly tenantId: string;
-  readonly consentRecordedAt: string;
-  readonly sampleCount: number;
-}
-
-export interface VoiceHealth {
-  readonly provider: string;
-  readonly ok: boolean;
-  readonly suggestion?: string;
-}
-
-export interface VoiceRealtimeOptions {
-  readonly root?: string;
-}
-
-export async function readVoiceDebug(root = process.cwd(), limit = 10): Promise<unknown[]> {
-  return readCapabilityDebug("voice-realtime", { root, limit });
-}
-
-export class VoiceRealtime {
-  readonly #root: string;
-
-  constructor(options: VoiceRealtimeOptions = {}) {
-    this.#root = options.root ?? process.cwd();
-  }
-
-  async startSession(input: { tenantId: string; threadId: string }): Promise<VoiceSession> {
-    const session: VoiceSession = {
-      id: assetId("voice_session", input.threadId),
-      tenantId: input.tenantId,
-      threadId: input.threadId,
-      room: `voice_${input.tenantId}_${input.threadId}`,
-      state: "listening",
-    };
-    await appendCapabilityDebug(
-      "voice-realtime",
-      { type: "session", session },
-      { root: this.#root },
-    );
-    return session;
-  }
-
-  async synthesizeNarration(
-    request: NarrationRequest,
-    brandInput: BrandContext | undefined,
-  ): Promise<VoiceAsset> {
-    const brand = requireBrandContext(brandInput, "voice-realtime");
-    const durationS = Math.max(
-      1,
-      Math.min(request.targetDurationS ?? Math.ceil(request.script.length / 24), 20),
-    );
-    const id = assetId("voice", `${brand.brandId}_${request.script}`);
-    const path = join(this.#root, ".nebutra", "generated", "voice-realtime", `${id}.wav`);
-    await mkdir(dirname(path), { recursive: true });
-    await writeFile(path, createToneWav(durationS, 330));
-    const asset: VoiceAsset = {
-      id,
-      tenantId: brand.tenantId,
-      kind: "voice",
-      path,
-      brandId: brand.brandId,
-      provider: "tone-local",
-      model: "brand-context-narration-v1",
-      createdAt: new Date().toISOString(),
-      license: { status: "commercial-ok", source: "deterministic local renderer" },
-      transcript: request.script,
-      durationS,
-      format: "wav",
-      voiceProfileId: request.voiceProfileId ?? "default",
-      metadata: { brandSource: brand.sourcePath },
-    };
-    await appendCapabilityDebug(
-      "voice-realtime",
-      { type: "narration", asset },
-      { root: this.#root },
-    );
-    return asset;
-  }
-
-  async enroll(input: {
-    tenantId: string;
-    samplePaths?: readonly string[];
-  }): Promise<VoiceProfile> {
-    const profile: VoiceProfile = {
-      id: assetId("voice_profile", input.tenantId),
-      tenantId: input.tenantId,
-      consentRecordedAt: new Date().toISOString(),
-      sampleCount: input.samplePaths?.length ?? 0,
-    };
-    await appendCapabilityDebug(
-      "voice-realtime",
-      { type: "enroll", profile },
-      { root: this.#root },
-    );
-    return profile;
-  }
-
-  async testMic(): Promise<VoiceHealth> {
-    return {
-      provider: "local-mic",
-      ok: Boolean(process.env.VOICE_MIC_PERMISSION === "granted"),
-      suggestion: "Grant microphone access in the desktop app before realtime sessions.",
-    };
-  }
-
-  async doctor(): Promise<VoiceHealth[]> {
-    return [
-      await this.testMic(),
-      {
-        provider: "voice-sidecar",
-        ok: Boolean(process.env.VOICE_SIDECAR_URL),
-        suggestion: "Set VOICE_SIDECAR_URL to enable realtime WebRTC/STT/TTS.",
-      },
-      {
-        provider: "remote-tts",
-        ok: Boolean(process.env.VOICE_REMOTE_API_KEY),
-        suggestion: "Set VOICE_REMOTE_API_KEY to enable remote voice fallback.",
-      },
-    ];
-  }
-}

package/tsconfig.json

@@ -1,12 +0,0 @@
-{
-  "extends": "../../../tsconfig.base.json",
-  "compilerOptions": {
-    "module": "ESNext",
-    "moduleResolution": "bundler",
-    "target": "esnext",
-    "types": ["node"],
-    "incremental": false
-  },
-  "include": ["src", "examples"],
-  "exclude": ["node_modules", "dist"]
-}

package/tsup.config.ts

@@ -1,16 +0,0 @@
-import { defineConfig } from "tsup";
-
-export default defineConfig({
-  entry: ["src/index.ts", "src/cli.ts"],
-  format: ["esm"],
-  dts: true,
-  sourcemap: true,
-  clean: true,
-  target: "es2022",
-  external: [
-    "@nebutra/audio-pipeline",
-    "@nebutra/capability-kit/debug",
-    "@nebutra/errors",
-    "@nebutra/generation-context",
-  ],
-});