Live_RC/tests/tts.test.js

import test from "node:test";
import assert from "node:assert/strict";
import { createRequire } from "node:module";
import { chmodSync, mkdirSync, mkdtempSync, readdirSync, rmSync, writeFileSync } from "node:fs";
import { join } from "node:path";
import { tmpdir } from "node:os";

const require = createRequire(import.meta.url);
const { createTts, listVoices, pickVoice, resolvePiperBinary } = require("../server/tts.js");
const { installVoices, installAll, VOICES } = require("../server/tts_install.js");

// Stand-in for execFile: records each call and writes the requested output file, so the
// module's queueing/caching can be tested without a real Piper install.
function fakeRun(calls, { fail = false, delayMs = 5 } = {}) {
  let active = 0;
  return (binary, args, _options, callback) => {
    active += 1;
    calls.push({ binary, args, maxParallel: active });
    let text = "";
    const stdin = { end: (value) => { text = value; } };
    setTimeout(() => {
      active -= 1;
      if (fail) return callback(new Error("piper crashed"));
      writeFileSync(args[args.indexOf("-f") + 1], `RIFF${text}`);
      return callback(null);
    }, delayMs);
    return { stdin };
  };
}

function setup() {
  const root = mkdtempSync(join(tmpdir(), "live-rc-tts-"));
  const voices = join(root, "voices");
  mkdirSync(voices);
  for (const name of ["sv_SE-nst-medium", "sv_SE-lisa-medium", "sv_SE-alma-low", "en_US-lessac-medium", "en_US-ryan-medium", "de_DE-thorsten-medium"]) {
    writeFileSync(join(voices, `${name}.onnx`), "");
    writeFileSync(join(voices, `${name}.onnx.json`), "{}");
  }
  writeFileSync(join(voices, "orphan-medium.onnx"), ""); // no .json config -> not a usable voice
  mkdirSync(join(root, "piper"));
  const binary = join(root, "piper", process.platform === "win32" ? "piper.exe" : "piper");
  writeFileSync(binary, "");
  chmodSync(binary, 0o755);
  return { root, voices, binary, cleanup: () => rmSync(root, { recursive: true, force: true }) };
}

test("voices need a matching .onnx.json, carry catalog gender labels, and the chosen voice wins", () => {
  const env = setup();
  try {
    const voices = listVoices([env.voices, join(env.root, "missing-dir")]);
    assert.deepEqual(voices.map((voice) => voice.name), ["de_DE-thorsten-medium", "en_US-lessac-medium", "en_US-ryan-medium", "sv_SE-alma-low", "sv_SE-lisa-medium", "sv_SE-nst-medium"]);
    const byName = Object.fromEntries(voices.map((voice) => [voice.name, voice]));
    assert.equal(byName["sv_SE-nst-medium"].gender, "male");
    assert.equal(byName["sv_SE-lisa-medium"].gender, "female");
    assert.equal(byName["en_US-ryan-medium"].gender, "male");
    assert.equal(byName["en_US-lessac-medium"].gender, "female");
    assert.equal(byName["sv_SE-alma-low"].gender, "", "a voice outside the catalog has no guessed gender");

    assert.equal(pickVoice(voices, "sv").name, "sv_SE-nst-medium", "Swedish defaults to the male NST voice");
    assert.equal(pickVoice(voices, "en-US").name, "en_US-ryan-medium", "English defaults to the male Ryan voice");
    assert.equal(pickVoice(voices, "sv", "sv_SE-lisa-medium").name, "sv_SE-lisa-medium", "the voice chosen in Settings is used");
    assert.equal(pickVoice(voices, "sv", "en_US-lessac-medium").name, "sv_SE-nst-medium", "a chosen voice in another language is ignored");
    assert.equal(pickVoice(voices, "sv", "sv_SE-removed-medium").name, "sv_SE-nst-medium", "an uninstalled choice falls back to the default");
    assert.equal(pickVoice(voices, "fi"), null);
  } finally { env.cleanup(); }
});

test("status offers the catalog voices that aren't installed yet, with gender and accent", () => {
  const env = setup();
  try {
    const status = createTts({ dataDir: env.root, projectDir: env.root, voiceDirs: [env.voices] }).status();
    const offered = Object.fromEntries(status.downloadable.map((voice) => [voice.name, voice]));
    assert.equal(offered["sv_SE-nst-medium"], undefined, "installed voices are not offered again");
    assert.deepEqual(offered["sv_SE-alma-medium"], { name: "sv_SE-alma-medium", language: "sv", gender: "female", accent: "", label: "Alma" });
    assert.equal(offered["en_GB-alan-medium"].accent, "gb");
    assert.equal(offered["en_GB-alan-medium"].gender, "male");
    assert.equal(status.voices.find((voice) => voice.name === "en_US-ryan-medium").accent, "us");
  } finally { env.cleanup(); }
});

test("a standard install skips the extra voices; picking one downloads only that voice", async () => {
  const env = setup();
  const originalFetch = globalThis.fetch;
  const urls = [];
  globalThis.fetch = async (url) => { urls.push(String(url)); return new Response("model"); };
  try {
    const target = join(env.root, "fresh-voices");
    await installVoices({ voicesDir: target, log: () => {} });
    const standard = VOICES.filter((voice) => !voice.extra).map((voice) => voice.name);
    assert.deepEqual(standard, ["sv_SE-nst-medium", "sv_SE-lisa-medium", "en_US-ryan-medium", "en_US-lessac-medium"]);
    assert.equal(urls.length, standard.length * 2, "model + config per standard voice");
    assert.ok(!urls.some((url) => url.includes("en_GB")), "no extra voice in a standard install");

    urls.length = 0;
    await installVoices({ voicesDir: target, log: () => {}, names: ["en_GB-alan-medium"] });
    assert.deepEqual(urls.map((url) => url.split("/main/")[1]), ["en/en_GB/alan/medium/en_GB-alan-medium.onnx.json", "en/en_GB/alan/medium/en_GB-alan-medium.onnx"]);
    assert.ok(readdirSync(target).includes("en_GB-alan-medium.onnx"));
    await assert.rejects(installAll({ projectDir: env.root, hasBinary: true, voice: "made_up-voice" }), /Unknown voice/);
  } finally {
    globalThis.fetch = originalFetch;
    env.cleanup();
  }
});

test("the Piper copy installed into the project is preferred over one on PATH", () => {
  const env = setup();
  try {
    assert.equal(resolvePiperBinary("", env.root), env.binary);
    assert.equal(resolvePiperBinary(env.binary, "/nonexistent"), env.binary, "an explicit path wins");
  } finally { env.cleanup(); }
});

test("each line is generated once, cached, deduplicated, and generated one at a time", async () => {
  const env = setup();
  const calls = [];
  try {
    const tts = createTts({ dataDir: join(env.root, "data"), projectDir: env.root, voiceDirs: [env.voices], run: fakeRun(calls) });
    const [a, b] = await Promise.all([
      tts.speech({ text: "Heat  ett startar.", language: "sv", rate: 1 }),
      tts.speech({ text: "Heat ett startar.", language: "sv", rate: 1 }),
    ]);
    assert.equal(a, b, "whitespace-normalized duplicates share one file");
    assert.equal(calls.length, 1, "concurrent identical requests only run Piper once");
    await tts.speech({ text: "Heat ett startar.", language: "sv", rate: 1 });
    assert.equal(calls.length, 1, "a cached line never runs Piper again");

    await Promise.all([tts.speech({ text: "One", language: "en" }), tts.speech({ text: "Two", language: "en" }), tts.speech({ text: "Tre", language: "sv", rate: 0.5 })]);
    assert.equal(calls.length, 4);
    assert.ok(calls.every((call) => call.maxParallel === 1), "generations never overlap");
    assert.ok(calls[1].args.includes(join(env.voices, "en_US-ryan-medium.onnx")), "the English voice is used for English text");
    await tts.speech({ text: "Heat ett startar.", language: "sv", rate: 1, voice: "sv_SE-lisa-medium" });
    assert.equal(calls.length, 5, "the same text in another voice is a separate cached line");
    assert.ok(calls[4].args.includes(join(env.voices, "sv_SE-lisa-medium.onnx")));
    assert.equal(calls[3].args[calls[3].args.indexOf("--length_scale") + 1], "1.6", "slow speech rate maps to a longer (clamped) length scale");

    const status = tts.status();
    assert.equal(status.available, true);
    assert.equal(status.cachedLines, 5);
    assert.equal(readdirSync(join(env.root, "data", "tts")).filter((file) => file.endsWith(".tmp")).length, 0, "no temp files are left behind");
  } finally { env.cleanup(); }
});

test("missing voice or failing Piper reports clear errors, and pre-generation counts failures", async () => {
  const env = setup();
  const calls = [];
  try {
    const tts = createTts({ dataDir: join(env.root, "data"), projectDir: env.root, voiceDirs: [env.voices], run: fakeRun(calls, { fail: true }) });
    await assert.rejects(tts.speech({ text: "Moi", language: "fi" }), (error) => error.code === "unavailable");
    await assert.rejects(tts.speech({ text: "   ", language: "sv" }), (error) => error.code === "bad_request");
    await assert.rejects(tts.speech({ text: "Hej", language: "sv" }), /piper crashed/);

    const result = await tts.pregenerate([{ text: "Hej", language: "sv" }, { text: "Hi", language: "en" }, { text: "", language: "sv" }]);
    assert.deepEqual(result, { running: false, total: 2, done: 2, failed: 2 });
  } finally { env.cleanup(); }
});