dsh-voice-mode 0.7.18 → 0.7.19

package/README.en.md

@@ -3,7 +3,7 @@
 [![License: MIT](https://img.shields.io/github/license/qishuilalala/dsh-voice-mode?style=flat-square&color=blue)](../../LICENSE)
 [![Latest Release](https://img.shields.io/github/v/release/qishuilalala/dsh-voice-mode?style=flat-square&color=brightgreen&include_prereleases)](https://github.com/qishuilalala/dsh-voice-mode/releases)
 [![npm version](https://img.shields.io/npm/v/dsh-voice-mode?style=flat-square&color=orange)](https://www.npmjs.com/package/dsh-voice-mode)
-[![Tests: 424 passing](https://img.shields.io/badge/tests-424%20%E2%9C%93-2ea043?style=flat-square)](../../docs/rules/STATE.md)
+[![Tests: 431 passing](https://img.shields.io/badge/tests-431%20%E2%9C%93-2ea043?style=flat-square)](../../docs/rules/STATE.md)
 
 Full-duplex voice conversation mode for DeepSeek Harness (dsh): speak, get a
 spoken answer. Streamed zipformer2 ASR → editable draft → auto send → the
@@ -171,6 +171,8 @@ Effect timing: `voice`/`rate`/`ttsEngine`/`kokoroModel`/`spokenFormat` take effe
 
 ### Common voices (full list: `node scripts/list-voices.mjs`)
 
+> **Desktop (Electron) note**: the official desktop host runs inside Electron, whose V8 forbids the external buffers returned by Kokoro's native add-on. The plugin automatically runs Kokoro in a real Node.js instead — **it needs Node.js >= 18 on PATH** (or set `DSHVM_NODE` to its path). Without it Kokoro reports an error; use the default Edge engine or local VITS (both work on desktop).
+
 | ShortName | Description |
 | --- | --- |
 | `zh-CN-XiaoxiaoNeural` | Xiaoxiao · Female (default) |

package/README.md

@@ -3,7 +3,7 @@
 [![License: MIT](https://img.shields.io/github/license/qishuilalala/dsh-voice-mode?style=flat-square&color=blue)](LICENSE)
 [![Latest Release](https://img.shields.io/github/v/release/qishuilalala/dsh-voice-mode?style=flat-square&color=brightgreen&include_prereleases)](https://github.com/qishuilalala/dsh-voice-mode/releases)
 [![npm version](https://img.shields.io/npm/v/dsh-voice-mode?style=flat-square&color=orange)](https://www.npmjs.com/package/dsh-voice-mode)
-[![Tests: 424 passing](https://img.shields.io/badge/tests-424%20%E2%9C%93-2ea043?style=flat-square)](../../docs/rules/STATE.md)
+[![Tests: 431 passing](https://img.shields.io/badge/tests-431%20%E2%9C%93-2ea043?style=flat-square)](../../docs/rules/STATE.md)
 
 DeepSeek Harness 语音双工对话模式：会话内一键进入 → 边说边出字的流式识别 → 停顿自动发送 → 最终答复按句流式朗读 + 实时字幕，开口即可打断（真 barge-in）。无需 API Key，识别模型在本地宿主端推理。
 
@@ -129,6 +129,8 @@ bundle 插件安装后需重启 dsh 生效（Linux：`systemctl restart dsh`；
 
 ### 本地音色（VITS / Kokoro）
 
+> **桌面端（Electron）提示**：官方桌面端的宿主运行在 Electron 里，其 V8 不允许 Kokoro 的原生 addon 返回外部缓冲区。插件会自动改用一个真正的 Node.js 来跑 Kokoro——**需要系统 PATH 上有 Node.js ≥18**（或用环境变量 `DSHVM_NODE` 指定路径）；没有则 Kokoro 会提示失败，请改用默认的 Edge 或本地 VITS（两者在桌面端不受影响）。
+
 - **VITS（纯中文）**：`suyingxue` 素映雪·女 / `gunian` 顾念·男 / `fushiyu` 傅斯遇·女 / `bingjiao` 冰娇·男 / `bazong` 霸总·男
 - **Kokoro（中英混读均可）**：103 个音色全量入表，面板按编号 + 实测性别标注；四个常用男声置顶：`62` 深沉 / `68` 浑厚 / `75` 清亮 / `76` 磁性；中文女声 `48` 小北 / `49` 小妮 / `50` 小小 / `51` 小艺。音色只是风格向量，**语言能力与音色无关**。
 

package/lib/client.js

@@ -3099,7 +3099,7 @@ var TELEMETRY_VIEW = [
   { stage: "first-tts-chunk", key: "telFirstChunk" },
   { stage: "first-audio-played", key: "telFirstPlayed" }
 ];
-var BUILD_TAG = "552d191";
+var BUILD_TAG = "62781d5";
 var TELEMETRY_FLAG = "dsh-voice-mode.telemetry";
 var telemetryEnabled = typeof localStorage !== "undefined" && localStorage.getItem(TELEMETRY_FLAG) === "1";
 console.log("[dsh-voice] build=" + BUILD_TAG);

package/lib/index.js

@@ -1,12 +1,11 @@
 // src/index.ts
 import z from "@deepseek-ai/schemastery";
-import { dirname as dirname3, join as join5 } from "node:path";
+import { dirname as dirname2, join as join5 } from "node:path";
 import { homedir as homedir2 } from "node:os";
 import { rm } from "node:fs/promises";
 
 // src/asr-host.ts
 import { statSync } from "node:fs";
-import { stat as stat2 } from "node:fs/promises";
 import { join as join2 } from "node:path";
 import { fileURLToPath } from "node:url";
 import { Worker } from "node:worker_threads";
@@ -185,7 +184,7 @@ async function sha256OfFile(path) {
   return hash.digest("hex");
 }
 async function ensureModelFile(opts) {
-  const { repo, repoDir, spec, primaryHost, allowCustomHost, broadcast } = opts;
+  const { repoDir, spec, primaryHost, broadcast } = opts;
   const localPath = join(repoDir, spec.file);
   const partPath = `${localPath}.part`;
   if ((await stat(localPath).catch(() => null))?.isFile()) {
@@ -605,8 +604,8 @@ function createAsrRuntime(options) {
     }
     let finMap = finalized.get(sessionId);
     const myGen = resetGen.get(sessionId) ?? 0;
-    const cached = finMap?.get(epoch);
-    if (cached !== void 0) return { text: cached };
+    const cached2 = finMap?.get(epoch);
+    if (cached2 !== void 0) return { text: cached2 };
     let sessSegs = segments.get(sessionId);
     if (!sessSegs) {
       sessSegs = /* @__PURE__ */ new Map();
@@ -847,10 +846,6 @@ function createAsrRuntime(options) {
       senseWorkerLangKey = "";
     },
     modelStatus: () => {
-      const statFile = async (dir, repo, name2) => {
-        const st = await stat2(join2(dir, repo, name2)).catch(() => null);
-        return { exists: !!st?.isFile(), size: st?.size ?? 0 };
-      };
       const asrFiles = MODEL_FILES.map((n) => ({
         name: n.file,
         exists: (() => {
@@ -1300,6 +1295,50 @@ var TtsQueue = class {
 
 // src/tts-local.ts
 import { fork } from "node:child_process";
+
+// src/tts-runtime.ts
+import { spawnSync } from "node:child_process";
+var probeNode = (command, env) => {
+  try {
+    const r = spawnSync(
+      command,
+      ["-p", "JSON.stringify([process.versions.node, !!process.versions.electron, process.execPath])"],
+      { env, encoding: "utf8", timeout: 5e3, windowsHide: true }
+    );
+    if (r.status !== 0) return null;
+    const [version, electron, execPath] = JSON.parse(r.stdout.trim());
+    return !electron && Number.parseInt(version, 10) >= 18 && typeof execPath === "string" ? execPath : null;
+  } catch {
+    return null;
+  }
+};
+var NATIVE_RUNTIME_UNAVAILABLE = 'Local Kokoro cannot run inside the Electron-based desktop host (native add-on blocked: "External buffers are not allowed"). Install Node.js >= 18 on PATH (or set DSHVM_NODE to its path), or switch the read-aloud engine to Edge or VITS.';
+var cached;
+function resolveNativeRuntime(opts = {}) {
+  const env = opts.env ?? process.env;
+  const useCache = opts.useCache ?? opts.env === void 0;
+  if (useCache && cached !== void 0) return cached;
+  const electron = opts.electron ?? !!process.versions.electron;
+  let result;
+  if (!electron) {
+    result = { env };
+  } else {
+    const childEnv = { ...env };
+    delete childEnv.ELECTRON_RUN_AS_NODE;
+    const probe = opts.probe ?? probeNode;
+    const candidates = [env.DSHVM_NODE, "node"].filter((c) => typeof c === "string" && c.length > 0);
+    let execPath = null;
+    for (const c of candidates) {
+      execPath = probe(c, childEnv);
+      if (execPath) break;
+    }
+    result = execPath ? { execPath, env: childEnv } : null;
+  }
+  if (useCache) cached = result;
+  return result;
+}
+
+// src/tts-local.ts
 import { fileURLToPath as fileURLToPath2 } from "node:url";
 import { join as join3 } from "node:path";
 import { statSync as statSync2 } from "node:fs";
@@ -1668,7 +1707,12 @@ function createSherpaLocalEngine(options) {
               if (!treeOk) throw new Error("local TTS model download failed: espeak-ng-data");
             }
             if (!child) {
-              child = fork(workerPath, [], { stdio: ["ignore", "ignore", "pipe", "ipc"] });
+              const runtime = options.kind === "kokoro" ? resolveNativeRuntime() : { env: process.env };
+              if (!runtime) throw new Error(NATIVE_RUNTIME_UNAVAILABLE);
+              child = fork(workerPath, [], {
+                stdio: ["ignore", "ignore", "pipe", "ipc"],
+                ...runtime.execPath ? { execPath: runtime.execPath, env: runtime.env } : {}
+              });
               let stderrTail = "";
               child.stderr?.on("data", (chunk) => {
                 stderrTail += String(chunk);
@@ -2143,7 +2187,7 @@ function previewSample(engine, voice) {
 // src/settings-store.ts
 import { chmodSync, mkdirSync, readFileSync, renameSync, writeFileSync } from "node:fs";
 import { homedir } from "node:os";
-import { dirname as dirname2, join as join4 } from "node:path";
+import { dirname, join as join4 } from "node:path";
 var SETTINGS_FILE_NAME = "voice-mode.settings.json";
 function settingsFilePath(profileHome) {
   const home = typeof profileHome === "string" && profileHome ? profileHome : process.env.DSH_HOME || join4(homedir(), ".dsh");
@@ -2181,7 +2225,7 @@ function readOverrides(file, keys) {
   }
 }
 function writeOverrides(file, values) {
-  mkdirSync(dirname2(file), { recursive: true });
+  mkdirSync(dirname(file), { recursive: true });
   const tmp = `${file}.${process.pid}.${Date.now()}.tmp`;
   writeFileSync(tmp, JSON.stringify(values, null, 2) + "\n", { encoding: "utf8", mode: 384 });
   try {
@@ -2536,7 +2580,7 @@ function apply(ctx, config) {
     if (loaded.warn) console.warn(`[dsh-voice-mode] ignoring settings overlay ${settingsFile}: ${loaded.warn}`);
     let migrated = false;
     if (!loaded.exists) {
-      const legacy = readLegacyVoiceSettings(dirname3(settingsFile), SETTING_KEYS);
+      const legacy = readLegacyVoiceSettings(dirname2(settingsFile), SETTING_KEYS);
       if (Object.keys(legacy).length > 0) {
         loaded.values = legacy;
         migrated = true;

package/package.json

@@ -2,7 +2,7 @@
   "name": "dsh-voice-mode",
   "description": "Full-duplex voice plugin for DeepSeek Harness: local zipformer2 streaming ASR (no API key) → editable draft; Edge TTS or local VITS / Kokoro read-aloud with live captions; true barge-in; hardened HTTP surface + model SHA256 pinning; compatible with all dsh releases since 0.1.1-rc.2 (verified 0.1.1-rc.2 → 0.2.0-rc.2).",
   "icon": "./icon.svg",
-  "version": "0.7.18",
+  "version": "0.7.19",
   "type": "module",
   "main": "lib/index.js",
   "repository": {
@@ -47,7 +47,7 @@
   "scripts": {
     "build": "node build.mjs",
     "prepack": "node build.mjs",
-    "test": "node test/segmenter.test.mjs && node test/aec.test.mjs && node test/download.test.mjs && node test/endpoint.test.mjs && node test/endpoint-short.test.mjs && node test/resample.test.mjs && node test/sense-worker.test.mjs && node test/detect-route.test.mjs && node test/verify-client.mjs && node test/asr-host-rebuild.test.mjs && node test/barge-in-manual.test.mjs && node test/sense-lang.test.mjs && node test/caption-a11y.test.mjs && node test/emotion.test.mjs && node test/emotion-tts-local.test.mjs && node test/emotion-integration.test.mjs && node test/backchannel.test.mjs && node test/strings-coverage.test.mjs && node test/yield-ms-wiring.test.mjs && node test/matchBackchannel.test.mjs && node test/settings-load.test.mjs && node test/preview-error.test.mjs && node test/hold-clear.test.mjs && node test/barge-in-detect.test.mjs && node test/wakeword.test.mjs && node test/wake-standby.test.mjs && node test/wake-flow.test.mjs && node test/tts-playback.test.mjs && node test/settings-store.test.mjs && node test/i18n.test.mjs && node test/i18n-protocol.test.mjs && node test/package-install.test.mjs",
+    "test": "node test/segmenter.test.mjs && node test/aec.test.mjs && node test/download.test.mjs && node test/endpoint.test.mjs && node test/endpoint-short.test.mjs && node test/resample.test.mjs && node test/sense-worker.test.mjs && node test/detect-route.test.mjs && node test/verify-client.mjs && node test/asr-host-rebuild.test.mjs && node test/barge-in-manual.test.mjs && node test/sense-lang.test.mjs && node test/caption-a11y.test.mjs && node test/emotion.test.mjs && node test/emotion-tts-local.test.mjs && node test/emotion-integration.test.mjs && node test/backchannel.test.mjs && node test/strings-coverage.test.mjs && node test/yield-ms-wiring.test.mjs && node test/matchBackchannel.test.mjs && node test/settings-load.test.mjs && node test/preview-error.test.mjs && node test/hold-clear.test.mjs && node test/barge-in-detect.test.mjs && node test/wakeword.test.mjs && node test/wake-standby.test.mjs && node test/wake-flow.test.mjs && node test/tts-playback.test.mjs && node test/settings-store.test.mjs && node test/i18n.test.mjs && node test/i18n-protocol.test.mjs && node test/package-install.test.mjs && node test/tts-runtime.test.mjs",
     "verify:client": "node test/verify-client.mjs",
     "prefetch": "node scripts/prefetch.mjs",
     "list-voices": "node scripts/list-voices.mjs",