feat: enhance TTS CLI with sox combining and wait flag

tts.ts

@@ -1,20 +1,51 @@
 import os from "node:os";
+import { parseArgs } from "node:util";
+import { $, which } from "bun";
 import chalk from "chalk";
 
+const { values: args } = parseArgs({
+  args: Bun.argv,
+  options: {
+    wait: { type: "boolean" },
+    w: { type: "boolean" },
+  },
+  strict: true,
+  allowPositionals: true,
+});
+
+const wait = args.wait ?? args.w;
 const text = await Bun.stdin.text();
 
-if (!text) process.exit(1);
+if (!text) {
+  console.error("No input given.");
+  process.exit(1);
+}
 
 const home = os.homedir();
+const cwd = `${home}/.ttsc/`;
+$.cwd(cwd);
+
+await $`rm *.wav`.nothrow();
+
+console.log(chalk.blue("Welcome to TTSC!"));
+console.log(
+  chalk.yellow(
+    "Beginning TTS processing... (This will take extra time the first time)",
+  ),
+);
 
 const koko = Bun.spawn({
   cmd: ["uv", "run", "--", "python", "-u", "main.py"],
-  cwd: `${home}/.ttsc/`,
+  cwd,
   stdin: Buffer.from(text),
   stdout: "pipe",
   stderr: "inherit",
 });
 
+if (!wait) {
+  koko.exited.then(() => console.log(chalk.green("\nProcessing finished.")));
+}
+
 const decoder = new TextDecoder();
 for await (const line of koko.stdout) {
   const decoded = decoder.decode(line).trim();
@@ -28,16 +59,40 @@ for await (const line of koko.stdout) {
   }
 
   for (const info of infos) {
-    console.log(`\n${chalk.blue(info.file)}: ${info.text}\n`);
+    const file = `${home}/.ttsc/${info.file}.wav`;
+    console.log(`\n${chalk.blue(file)}:\n${info.text}`);
 
+    if (!wait) {
       await Bun.spawn({
-      cmd: ["mpv", `${info.file}.wav`],
+        cmd: ["mpv", file],
-      cwd: `${home}/.ttsc/`,
+        cwd,
         stdin: "ignore",
         stdout: "inherit",
         stderr: "inherit",
       }).exited;
     }
+  }
 }
 
 await koko.exited;
+if (!wait) process.exit();
+
+console.log(chalk.green("\nProcessing finished."));
+
+console.log(chalk.yellow(`Combining audio with sox...`));
+if (which("sox")) {
+  await $`sox *.wav output.wav`;
+  console.log(chalk.green(`Playing audio from ${cwd}output.wav`));
+} else {
+  console.error("Sox not found. Unable to combine outputs.");
+  console.log(chalk.yellow(`Audio available in ${cwd}`));
+  process.exit();
+}
+
+await Bun.spawn({
+  cmd: ["mpv", "output.wav"],
+  cwd,
+  stdin: "ignore",
+  stdout: "inherit",
+  stderr: "inherit",
+}).exited;

tts/main.py

@@ -9,5 +9,5 @@ pipeline = KPipeline(lang_code="a", repo_id="hexgrad/Kokoro-82M")
 generator = pipeline(text, voice="af_heart", speed=1.2)
 
 for i, (gs, ps, audio) in enumerate(generator):
-    print(json.dumps({"file": i, "text": gs}))
     sf.write(f"{i}.wav", audio, 24000)
+    print(json.dumps({"file": i, "text": gs}))