migrate to root dir

This commit is contained in:
2025-08-08 19:26:21 +02:00
parent cf8219691b
commit 8720500442
41 changed files with 2478 additions and 4440 deletions

45
.gitignore vendored
View File

@@ -1,8 +1,45 @@
# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
node_modules/ # dependencies
/node_modules
/.pnp
.pnp.*
.yarn/*
!.yarn/patches
!.yarn/plugins
!.yarn/releases
!.yarn/versions
.env # testing
/coverage
stories/sample_story/ # next.js
/.next/
/out/
dist/ # production
/build
# misc
.DS_Store
*.pem
# debug
npm-debug.log*
yarn-debug.log*
yarn-error.log*
.pnpm-debug.log*
# env files (can opt-in for committing if needed)
.env*
# vercel
.vercel
# typescript
*.tsbuildinfo
next-env.d.ts
stories/sample_story/

View File

@@ -1 +1,36 @@
# audiobooks-hustle This is a [Next.js](https://nextjs.org) project bootstrapped with [`create-next-app`](https://nextjs.org/docs/app/api-reference/cli/create-next-app).
## Getting Started
First, run the development server:
```bash
npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev
```
Open [http://localhost:3000](http://localhost:3000) with your browser to see the result.
You can start editing the page by modifying `app/page.tsx`. The page auto-updates as you edit the file.
This project uses [`next/font`](https://nextjs.org/docs/app/building-your-application/optimizing/fonts) to automatically optimize and load [Geist](https://vercel.com/font), a new font family for Vercel.
## Learn More
To learn more about Next.js, take a look at the following resources:
- [Next.js Documentation](https://nextjs.org/docs) - learn about Next.js features and API.
- [Learn Next.js](https://nextjs.org/learn) - an interactive Next.js tutorial.
You can check out [the Next.js GitHub repository](https://github.com/vercel/next.js) - your feedback and contributions are welcome!
## Deploy on Vercel
The easiest way to deploy your Next.js app is to use the [Vercel Platform](https://vercel.com/new?utm_medium=default-template&filter=next.js&utm_source=create-next-app&utm_campaign=create-next-app-readme) from the creators of Next.js.
Check out our [Next.js deployment documentation](https://nextjs.org/docs/app/building-your-application/deploying) for more details.

43
app/api/run/route.ts Normal file
View File

@@ -0,0 +1,43 @@
import path from "path";
import { runStoryPipeline } from "@/lib/pipeline/pipeline";
export const runtime = "nodejs";
export const dynamic = "force-dynamic";
export const maxDuration = 300; // seconds
type RunBody = {
storyName: string;
force?: boolean;
skipUpload?: boolean;
concurrency?: number;
};
function isRunBody(value: unknown): value is RunBody {
if (typeof value !== "object" || value === null) return false;
const v = value as Record<string, unknown>;
return typeof v.storyName === "string";
}
export async function POST(request: Request) {
const bodyUnknown: unknown = await request.json().catch(() => null);
if (!isRunBody(bodyUnknown)) {
return Response.json({ ok: false, error: "'storyName' is required" }, { status: 400 });
}
const storyName = bodyUnknown.storyName;
const force = !!bodyUnknown.force;
const skipUpload = !!bodyUnknown.skipUpload;
const concurrency = Math.max(1, Math.min(10, Number(bodyUnknown.concurrency) || 3));
// Keep using repo root for stories unless you move them into the Next.js folder
const baseDir = path.resolve(process.cwd(), "..");
try {
const result = await runStoryPipeline(storyName, { force, skipUpload, concurrency, baseDir });
return Response.json({ ok: true, result });
} catch (error: unknown) {
const message = error instanceof Error ? error.message : String(error);
return Response.json({ ok: false, error: message }, { status: 500 });
}
}

View File

Before

Width:  |  Height:  |  Size: 25 KiB

After

Width:  |  Height:  |  Size: 25 KiB

139
app/page.tsx Normal file
View File

@@ -0,0 +1,139 @@
"use client";
import { useState } from "react";
type PipelineResult = {
videoPath: string;
finalAudioPath: string;
imageFiles: string[];
};
type ApiSuccess = { ok: true; result: PipelineResult };
type ApiError = { ok: false; error?: string };
type ApiResponse = ApiSuccess | ApiError;
function isApiResponse(value: unknown): value is ApiResponse {
if (typeof value !== "object" || value === null) return false;
const v = value as Record<string, unknown>;
if (v.ok === true) {
const r = v.result as Record<string, unknown> | undefined;
return (
typeof r === "object" &&
r !== null &&
typeof r.videoPath === "string" &&
typeof r.finalAudioPath === "string" &&
Array.isArray(r.imageFiles)
);
}
return v.ok === false;
}
export default function Home() {
const [storyName, setStoryName] = useState("sample_story");
const [concurrency, setConcurrency] = useState(3);
const [force, setForce] = useState(false);
const [skipUpload, setSkipUpload] = useState(true);
const [loading, setLoading] = useState(false);
const [result, setResult] = useState<PipelineResult | null>(null);
const [error, setError] = useState<string | null>(null);
async function runPipeline(e: React.FormEvent) {
e.preventDefault();
setLoading(true);
setError(null);
setResult(null);
try {
const res = await fetch("/api/run", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({ storyName, concurrency, force, skipUpload }),
});
const data: unknown = await res.json();
if (!res.ok || !isApiResponse(data) || data.ok !== true) {
const msg = isApiResponse(data) && data.ok === false ? data.error : undefined;
throw new Error(msg || `Request failed with status ${res.status}`);
}
setResult(data.result);
} catch (err: unknown) {
const message = err instanceof Error ? err.message : String(err);
setError(message);
} finally {
setLoading(false);
}
}
return (
<div className="min-h-screen w-full flex items-center justify-center p-8">
<div className="w-full max-w-2xl space-y-6">
<h1 className="text-2xl font-semibold">Audiobooks Hustle Run Pipeline</h1>
<form onSubmit={runPipeline} className="space-y-4">
<div>
<label className="block text-sm font-medium mb-1">Story name</label>
<input
className="w-full rounded-md border px-3 py-2"
placeholder="sample_story"
value={storyName}
onChange={(e) => setStoryName(e.target.value)}
required
/>
<p className="text-xs text-muted-foreground mt-1">
Must match a folder in the repo root at <code>stories/&lt;storyName&gt;</code> containing
<code>source.txt</code> and <code>config.yaml</code>.
</p>
</div>
<div className="grid grid-cols-2 gap-4">
<div>
<label className="block text-sm font-medium mb-1">Concurrency</label>
<input
type="number"
min={1}
max={10}
className="w-full rounded-md border px-3 py-2"
value={concurrency}
onChange={(e) => setConcurrency(parseInt(e.target.value || "3", 10))}
/>
</div>
<div className="flex items-center gap-4">
<label className="inline-flex items-center gap-2">
<input type="checkbox" checked={force} onChange={(e) => setForce(e.target.checked)} />
<span>Force regenerate</span>
</label>
<label className="inline-flex items-center gap-2">
<input type="checkbox" checked={skipUpload} onChange={(e) => setSkipUpload(e.target.checked)} />
<span>Skip upload</span>
</label>
</div>
</div>
<button
type="submit"
className="rounded-md bg-black text-white px-4 py-2 disabled:opacity-50"
disabled={loading}
>
{loading ? "Running..." : "Run pipeline"}
</button>
</form>
{error && <div className="rounded-md border border-red-300 bg-red-50 p-3 text-sm text-red-700">{error}</div>}
{result && (
<div className="space-y-2 text-sm">
<div className="rounded-md border p-3">
<div className="font-medium">Video created:</div>
<pre className="overflow-auto whitespace-pre-wrap">{result.videoPath}</pre>
</div>
<div className="rounded-md border p-3">
<div className="font-medium">Final audio:</div>
<pre className="overflow-auto whitespace-pre-wrap">{result.finalAudioPath}</pre>
</div>
<div className="rounded-md border p-3">
<div className="font-medium">Images ({result.imageFiles?.length ?? 0}):</div>
<pre className="overflow-auto whitespace-pre-wrap">{(result.imageFiles || []).join("\n")}</pre>
</div>
</div>
)}
</div>
</div>
);
}

View File

@@ -2,22 +2,22 @@ import { spawn } from "child_process";
import { StoryConfig } from "./config"; import { StoryConfig } from "./config";
import * as path from "path"; import * as path from "path";
import * as fs from "fs"; import * as fs from "fs";
const ffmpeg = require("ffmpeg-static"); import ffmpeg from "ffmpeg-static";
const ffprobe = require("ffprobe-static"); import ffprobe from "ffprobe-static";
export function getDuration(file: string): Promise<number> { export function getDuration(file: string): Promise<number> {
const ffprobePath = ffprobe.path; const ffprobePath: string = (ffprobe as unknown as { path: string }).path;
const args = ["-v", "error", "-show_entries", "format=duration", "-of", "default=noprint_wrappers=1:nokey=1", file]; const args = ["-v", "error", "-show_entries", "format=duration", "-of", "default=noprint_wrappers=1:nokey=1", file];
const ffprobeProcess = spawn(ffprobePath, args); const ffprobeProcess = spawn(ffprobePath, args);
let duration = ""; let duration = "";
ffprobeProcess.stdout.on("data", (data: any) => { ffprobeProcess.stdout.on("data", (data: Buffer) => {
duration += data.toString(); duration += data.toString();
}); });
return new Promise<number>((resolve, reject) => { return new Promise<number>((resolve, reject) => {
ffprobeProcess.on("close", (code: any) => { ffprobeProcess.on("close", (code: number | null) => {
if (code === 0) { if (code === 0) {
resolve(parseFloat(duration)); resolve(parseFloat(duration));
} else { } else {
@@ -32,21 +32,13 @@ export async function getChunkDuration(chunkPath: string): Promise<number> {
} }
export async function generateSilence(duration: number, outputFile: string): Promise<void> { export async function generateSilence(duration: number, outputFile: string): Promise<void> {
const ffmpegPath = ffmpeg; const ffmpegPath: string = (ffmpeg as unknown as string) || "ffmpeg";
const args = ["-y", "-f", "lavfi", "-i", `anullsrc=r=44100:cl=stereo:d=${duration}`, outputFile]; const args = ["-y", "-f", "lavfi", "-i", `anullsrc=r=44100:cl=stereo:d=${duration}`, outputFile];
const ffmpegProcess = spawn(ffmpegPath, args); const ffmpegProcess = spawn(ffmpegPath, args);
ffmpegProcess.stdout.on("data", (data: any) => {
console.log(`stdout: ${data}`);
});
ffmpegProcess.stderr.on("data", (data: any) => {
console.error(`stderr: ${data}`);
});
return new Promise<void>((resolve, reject) => { return new Promise<void>((resolve, reject) => {
ffmpegProcess.on("close", (code: any) => { ffmpegProcess.on("close", (code: number | null) => {
if (code === 0) { if (code === 0) {
resolve(); resolve();
} else { } else {
@@ -57,6 +49,7 @@ export async function generateSilence(duration: number, outputFile: string): Pro
} }
export async function combineAudio(storyName: string, storyConfig: StoryConfig, audioFiles: string[]): Promise<void> { export async function combineAudio(storyName: string, storyConfig: StoryConfig, audioFiles: string[]): Promise<void> {
const ffmpegPath: string = (ffmpeg as unknown as string) || "ffmpeg";
const introFile = path.resolve("stories", storyName, storyConfig.config.intro_audio_file); const introFile = path.resolve("stories", storyName, storyConfig.config.intro_audio_file);
const outroFile = path.resolve("stories", storyName, storyConfig.config.outro_audio_file); const outroFile = path.resolve("stories", storyName, storyConfig.config.outro_audio_file);
const finalAudioDir = path.resolve("stories", storyName, "final_audio"); const finalAudioDir = path.resolve("stories", storyName, "final_audio");
@@ -66,18 +59,17 @@ export async function combineAudio(storyName: string, storyConfig: StoryConfig,
fs.mkdirSync(finalAudioDir, { recursive: true }); fs.mkdirSync(finalAudioDir, { recursive: true });
// First, concatenate the main audio files
const allFiles = [introFile, ...audioFiles.map((f) => path.resolve(f)), outroFile]; const allFiles = [introFile, ...audioFiles.map((f) => path.resolve(f)), outroFile];
const fileList = allFiles.map((f) => `file '${f.replace(/'/g, "'\\''")}'`).join("\n"); const fileList = allFiles.map((f) => `file '${f.replace(/'/g, "'\\''")}'`).join("\n");
const listFile = path.resolve("stories", storyName, "filelist.txt"); const listFile = path.resolve("stories", storyName, "filelist.txt");
require("fs").writeFileSync(listFile, fileList); fs.writeFileSync(listFile, fileList);
const concatArgs = ["-y", "-f", "concat", "-safe", "0", "-i", listFile, "-c", "copy", tempAudioFile]; const concatArgs = ["-y", "-f", "concat", "-safe", "0", "-i", listFile, "-c", "copy", tempAudioFile];
const concatProcess = spawn(ffmpeg, concatArgs); const concatProcess = spawn(ffmpegPath, concatArgs);
await new Promise<void>((resolve, reject) => { await new Promise<void>((resolve, reject) => {
concatProcess.on("close", (code: any) => { concatProcess.on("close", (code: number | null) => {
if (code === 0) { if (code === 0) {
resolve(); resolve();
} else { } else {
@@ -86,20 +78,19 @@ export async function combineAudio(storyName: string, storyConfig: StoryConfig,
}); });
}); });
// Then, get the duration of the concatenated audio
const duration = await getDuration(tempAudioFile); const duration = await getDuration(tempAudioFile);
if (!fs.existsSync(backgroundMusicFile)) { if (!fs.existsSync(backgroundMusicFile)) {
// If background music is missing, just copy the narration
await new Promise<void>((resolve, reject) => { await new Promise<void>((resolve, reject) => {
const args = ["-y", "-i", tempAudioFile, "-c:a", "libmp3lame", "-q:a", "4", finalAudioFile]; const args = ["-y", "-i", tempAudioFile, "-c:a", "libmp3lame", "-q:a", "4", finalAudioFile];
const p = spawn(ffmpeg, args); const p = spawn(ffmpegPath, args);
p.on("close", (code: any) => (code === 0 ? resolve() : reject(new Error(`ffmpeg copy failed ${code}`)))); p.on("close", (code: number | null) =>
code === 0 ? resolve() : reject(new Error(`ffmpeg copy failed ${code}`))
);
}); });
return; return;
} }
// Finally, mix the main audio with the looped/trimmed background music at a lower volume
const bgVolume = "0.2"; const bgVolume = "0.2";
const mixArgs = [ const mixArgs = [
"-y", "-y",
@@ -120,10 +111,10 @@ export async function combineAudio(storyName: string, storyConfig: StoryConfig,
finalAudioFile, finalAudioFile,
]; ];
const mixProcess = spawn(ffmpeg, mixArgs); const mixProcess = spawn(ffmpegPath, mixArgs);
return new Promise<void>((resolve, reject) => { return new Promise<void>((resolve, reject) => {
mixProcess.on("close", (code: any) => { mixProcess.on("close", (code: number | null) => {
if (code === 0) { if (code === 0) {
resolve(); resolve();
} else { } else {

View File

@@ -45,7 +45,10 @@ const StoryConfigSchema = z.object({
export_settings: z export_settings: z
.object({ .object({
format: z.string().optional().default("mp4"), format: z.string().optional().default("mp4"),
resolution: z.string().regex(/^\d+x\d+$/).default("1024x1024"), resolution: z
.string()
.regex(/^\d+x\d+$/)
.default("1024x1024"),
}) })
.default({ format: "mp4", resolution: "1024x1024" }), .default({ format: "mp4", resolution: "1024x1024" }),
}), }),

View File

@@ -3,11 +3,15 @@ import * as fs from "fs";
import * as path from "path"; import * as path from "path";
import { StoryConfig } from "./config"; import { StoryConfig } from "./config";
const openai = new OpenAI({ let openaiClient: OpenAI | null = null;
apiKey: process.env.OPENAI_API_KEY, function getOpenAI(): OpenAI {
}); if (!openaiClient) {
openaiClient = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
}
return openaiClient;
}
const allowedSizes = [ const allowedSizesValues = [
"256x256", "256x256",
"512x512", "512x512",
"1024x1024", "1024x1024",
@@ -16,10 +20,9 @@ const allowedSizes = [
"1792x1024", "1792x1024",
"1024x1792", "1024x1792",
] as const; ] as const;
type AllowedSize = (typeof allowedSizes)[number]; type AllowedSize = (typeof allowedSizesValues)[number];
function pickImageSize(resolution?: string): AllowedSize { function pickImageSize(resolution?: string): AllowedSize {
// Default square for simplicity
if (!resolution) return "1024x1024"; if (!resolution) return "1024x1024";
const match = resolution.match(/^(\d+)x(\d+)$/); const match = resolution.match(/^(\d+)x(\d+)$/);
if (!match) return "1024x1024"; if (!match) return "1024x1024";
@@ -47,7 +50,7 @@ Illustration for the following passage:
const size = pickImageSize(storyConfig.config.export_settings?.resolution); const size = pickImageSize(storyConfig.config.export_settings?.resolution);
const response = await openai.images.generate({ const response = await getOpenAI().images.generate({
model: "dall-e-3", model: "dall-e-3",
prompt, prompt,
n: 1, n: 1,

151
lib/pipeline/pipeline.ts Normal file
View File

@@ -0,0 +1,151 @@
import { loadStoryConfig } from "./config";
import { validatePublicDomain } from "./validator";
import { sanitizeText } from "./sanitizer";
import { chunkText } from "./chunker";
import { generateAudio, generateSingleAudio } from "./tts";
import { combineAudio, getChunkDuration } from "./audio";
import { generateImage } from "./images";
import { createVideo } from "./video";
import { createSrt } from "./subtitles";
import { generateYouTubeMetadata, uploadToYouTube, YouTubeMetadata } from "./uploader";
import * as path from "path";
import * as fs from "fs";
export interface RunPipelineOptions {
force?: boolean;
skipUpload?: boolean;
concurrency?: number;
baseDir?: string;
}
export interface RunPipelineResult {
storyName: string;
audioFiles: string[];
imageFiles: string[];
srtPath: string;
finalAudioPath: string;
videoPath: string;
metadata: YouTubeMetadata;
}
async function mapWithConcurrency<T>(
items: T[],
limit: number,
mapper: (item: T, index: number) => Promise<void>
): Promise<void> {
if (items.length === 0) return;
let nextIndex = 0;
const inFlight: Promise<void>[] = [];
const launchNext = () => {
if (nextIndex >= items.length) return;
const current = nextIndex++;
const p = mapper(items[current], current).finally(() => {
const idx = inFlight.indexOf(p);
if (idx >= 0) inFlight.splice(idx, 1);
});
inFlight.push(p);
};
for (let i = 0; i < Math.min(limit, items.length); i++) {
launchNext();
}
while (inFlight.length > 0 || nextIndex < items.length) {
while (inFlight.length < limit && nextIndex < items.length) {
launchNext();
}
await Promise.race(inFlight);
}
}
export async function runStoryPipeline(
storyName: string,
options: RunPipelineOptions = {}
): Promise<RunPipelineResult> {
const force = !!options.force;
const skipUpload = !!options.skipUpload;
const concurrency = Math.max(1, options.concurrency ?? 3);
const originalCwd = process.cwd();
const targetCwd = options.baseDir || originalCwd;
if (targetCwd && targetCwd !== originalCwd) {
process.chdir(targetCwd);
}
try {
const storyConfig = loadStoryConfig(storyName);
const validationResult = validatePublicDomain(storyConfig);
if (!validationResult.is_public_domain) {
throw new Error(validationResult.message);
}
const storyRoot = path.resolve("stories", storyName);
["audio", "images", "final_audio", "video"].forEach((d) =>
fs.mkdirSync(path.join(storyRoot, d), { recursive: true })
);
const sanitizedText = sanitizeText(storyName);
const chunks = chunkText(sanitizedText, storyConfig.config.chunk_size);
const introFile = path.join("stories", storyName, storyConfig.config.intro_audio_file);
const outroFile = path.join("stories", storyName, storyConfig.config.outro_audio_file);
if (!fs.existsSync(introFile) || force) {
await generateSingleAudio(storyConfig, "This is the intro.", introFile);
}
if (!fs.existsSync(outroFile) || force) {
await generateSingleAudio(storyConfig, "This is the outro.", outroFile);
}
const audioFiles: string[] = new Array(chunks.length);
const chunkDurations: number[] = new Array(chunks.length);
await mapWithConcurrency(chunks, concurrency, async (chunk, i) => {
const audioPath = path.join("stories", storyName, "audio", `chunk_${i}.mp3`);
if (!fs.existsSync(audioPath) || force) {
await generateAudio(storyConfig, storyName, chunk, i);
}
const duration = await getChunkDuration(audioPath);
audioFiles[i] = audioPath;
chunkDurations[i] = duration;
});
const imageFiles: string[] = new Array(chunks.length);
await mapWithConcurrency(chunks, concurrency, async (chunk, i) => {
const imagePath = path.join("stories", storyName, "images", `chunk_${i}_img0.png`);
if (!fs.existsSync(imagePath) || force) {
const generated = await generateImage(storyName, storyConfig, chunk, i, 0);
imageFiles[i] = generated;
} else {
imageFiles[i] = imagePath;
}
});
const srtPath = createSrt(storyName, chunks, chunkDurations);
await combineAudio(storyName, storyConfig, audioFiles);
const finalAudioPath = path.resolve("stories", storyName, "final_audio", "final.mp3");
await createVideo(storyName, storyConfig, imageFiles, chunkDurations, srtPath);
const videoPath = path.resolve("stories", storyName, "video", "final.mp4");
const metadata = generateYouTubeMetadata(storyConfig);
if (!skipUpload) {
await uploadToYouTube(videoPath, metadata);
}
return {
storyName,
audioFiles,
imageFiles,
srtPath,
finalAudioPath,
videoPath,
metadata,
};
} finally {
if (targetCwd !== originalCwd) {
process.chdir(originalCwd);
}
}
}

View File

@@ -1,7 +1,7 @@
import * as fs from "fs"; import * as fs from "fs";
import * as path from "path"; import * as path from "path";
const FORBIDDEN_WORDS = ["darn", "heck", "gosh"]; // Example list const FORBIDDEN_WORDS = ["darn", "heck", "gosh"];
export function sanitizeText(storyName: string): string { export function sanitizeText(storyName: string): string {
const sourcePath = path.join("stories", storyName, "source.txt"); const sourcePath = path.join("stories", storyName, "source.txt");

View File

@@ -3,9 +3,13 @@ import * as fs from "fs";
import * as path from "path"; import * as path from "path";
import { StoryConfig } from "./config"; import { StoryConfig } from "./config";
const openai = new OpenAI({ let openaiClient: OpenAI | null = null;
apiKey: process.env.OPENAI_API_KEY, function getOpenAI(): OpenAI {
}); if (!openaiClient) {
openaiClient = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
}
return openaiClient;
}
export async function generateAudio( export async function generateAudio(
storyConfig: StoryConfig, storyConfig: StoryConfig,
@@ -15,9 +19,9 @@ export async function generateAudio(
): Promise<string> { ): Promise<string> {
const speechFile = path.join("stories", storyName, "audio", `chunk_${index}.mp3`); const speechFile = path.join("stories", storyName, "audio", `chunk_${index}.mp3`);
const mp3 = await openai.audio.speech.create({ const mp3 = await getOpenAI().audio.speech.create({
model: "gpt-4o-mini-tts", model: "gpt-4o-mini-tts",
voice: storyConfig.config.tts_voice_id as any, voice: storyConfig.config.tts_voice_id as unknown as string,
input: chunk, input: chunk,
instructions: storyConfig.config.tts_instructions, instructions: storyConfig.config.tts_instructions,
}); });
@@ -29,9 +33,9 @@ export async function generateAudio(
} }
export async function generateSingleAudio(storyConfig: StoryConfig, text: string, outputFile: string): Promise<string> { export async function generateSingleAudio(storyConfig: StoryConfig, text: string, outputFile: string): Promise<string> {
const mp3 = await openai.audio.speech.create({ const mp3 = await getOpenAI().audio.speech.create({
model: "gpt-4o-mini-tts", model: "gpt-4o-mini-tts",
voice: storyConfig.config.tts_voice_id as any, voice: storyConfig.config.tts_voice_id as unknown as string,
input: text, input: text,
instructions: storyConfig.config.tts_instructions, instructions: storyConfig.config.tts_instructions,
}); });

View File

@@ -1,5 +1,4 @@
import { StoryConfig } from "./config"; import { StoryConfig } from "./config";
import { google } from "googleapis";
export interface YouTubeMetadata { export interface YouTubeMetadata {
title: string; title: string;

View File

@@ -20,7 +20,7 @@ export function validatePublicDomain(storyConfig: StoryConfig): ValidationResult
if (publicationYear > cutoffYear) { if (publicationYear > cutoffYear) {
return { return {
is_public_domain: false, is_public_domain: false,
message: `Work published in ${publicationYear} is not yet in the US public domain (cutoff: \u2264 ${cutoffYear}).`, message: `Work published in ${publicationYear} is not yet in the US public domain (cutoff: ${cutoffYear}).`,
}; };
} }

View File

@@ -1,14 +1,10 @@
import { spawn } from "child_process"; import { spawn } from "child_process";
import { StoryConfig } from "./config"; import { StoryConfig } from "./config";
import * as path from "path"; import * as path from "path";
const ffmpeg = require("ffmpeg-static"); import ffmpeg from "ffmpeg-static";
function escapeForFilter(filePath: string): string { function escapeForFilter(filePath: string): string {
return filePath return filePath.replace(/\\/g, "\\\\").replace(/:/g, "\\:").replace(/,/g, "\\,").replace(/'/g, "\\'");
.replace(/\\/g, "\\\\")
.replace(/:/g, "\\:")
.replace(/,/g, "\\,")
.replace(/'/g, "\\'");
} }
export async function createVideo( export async function createVideo(
@@ -18,6 +14,7 @@ export async function createVideo(
chunkDurations: number[], chunkDurations: number[],
srtPath: string srtPath: string
): Promise<void> { ): Promise<void> {
const ffmpegPath: string = (ffmpeg as unknown as string) || "ffmpeg";
const audioPath = path.resolve("stories", storyName, "final_audio", "final.mp3"); const audioPath = path.resolve("stories", storyName, "final_audio", "final.mp3");
const videoPath = path.resolve("stories", storyName, "video", "final.mp4"); const videoPath = path.resolve("stories", storyName, "video", "final.mp4");
const totalDuration = chunkDurations.reduce((a, b) => a + b, 0); const totalDuration = chunkDurations.reduce((a, b) => a + b, 0);
@@ -63,14 +60,10 @@ export async function createVideo(
videoPath, videoPath,
]; ];
const ffmpegProcess = spawn(ffmpeg, args); const ffmpegProcess = spawn(ffmpegPath, args);
ffmpegProcess.stdout.on("data", (data: any) => {});
ffmpegProcess.stderr.on("data", (data: any) => {});
return new Promise<void>((resolve, reject) => { return new Promise<void>((resolve, reject) => {
ffmpegProcess.on("close", (code: any) => { ffmpegProcess.on("close", (code: number | null) => {
if (code === 0) { if (code === 0) {
resolve(); resolve();
} else { } else {

16
next.config.ts Normal file
View File

@@ -0,0 +1,16 @@
import type { NextConfig } from "next";
const nextConfig: NextConfig = {
serverExternalPackages: [
"ffmpeg-static",
"ffprobe-static",
"googleapis",
"openai",
"js-yaml",
"p-limit",
"zod",
"dotenv",
],
};
export default nextConfig;

View File

@@ -1,17 +1,21 @@
{ {
"name": "project-noctivus", "name": "project-noctivus",
"version": "1.0.0", "version": "0.1.0",
"description": "An orchestrator for creating audiobooks from text.", "private": true,
"main": "dist/orchestrator.js",
"scripts": { "scripts": {
"start": "ts-node src/orchestrator.ts", "dev": "next dev --turbopack",
"build": "tsc -p tsconfig.json", "build": "next build",
"start:prod": "node dist/orchestrator.js", "start": "next start",
"lint": "eslint 'src/**/*.{ts,tsx}' --max-warnings=0", "lint": "next lint"
"format": "prettier --write .",
"preinstall": "npx only-allow pnpm"
}, },
"dependencies": { "dependencies": {
"class-variance-authority": "^0.7.1",
"clsx": "^2.1.1",
"lucide-react": "^0.539.0",
"next": "15.4.6",
"react": "19.1.0",
"react-dom": "19.1.0",
"tailwind-merge": "^3.3.1",
"dotenv": "^16.4.5", "dotenv": "^16.4.5",
"ffmpeg-static": "^5.2.0", "ffmpeg-static": "^5.2.0",
"ffprobe-static": "^3.1.0", "ffprobe-static": "^3.1.0",
@@ -22,16 +26,17 @@
"zod": "^3.23.8" "zod": "^3.23.8"
}, },
"devDependencies": { "devDependencies": {
"@types/js-yaml": "^4.0.9", "@eslint/eslintrc": "^3",
"@types/node": "^20.12.12", "@tailwindcss/postcss": "^4",
"eslint": "^9.7.0", "@types/node": "^20",
"eslint-config-prettier": "^9.1.0", "@types/react": "^19",
"eslint-plugin-import": "^2.29.1", "@types/react-dom": "^19",
"prettier": "^3.3.3", "eslint": "^9",
"ts-node": "^10.9.2", "eslint-config-next": "15.4.6",
"typescript": "^5.4.5" "eslint-plugin-react-hooks": "^5.1.0",
}, "eslint-plugin-react-refresh": "^0.4.13",
"engines": { "tailwindcss": "^4",
"node": ">=18.18" "tw-animate-css": "^1.3.6",
"typescript": "^5"
} }
} }

2079
pnpm-lock.yaml generated

File diff suppressed because it is too large Load Diff

2
pnpm-workspace.yaml Normal file
View File

@@ -0,0 +1,2 @@
onlyBuiltDependencies:
- ffmpeg-static

View File

@@ -1,41 +0,0 @@
# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
# dependencies
/node_modules
/.pnp
.pnp.*
.yarn/*
!.yarn/patches
!.yarn/plugins
!.yarn/releases
!.yarn/versions
# testing
/coverage
# next.js
/.next/
/out/
# production
/build
# misc
.DS_Store
*.pem
# debug
npm-debug.log*
yarn-debug.log*
yarn-error.log*
.pnpm-debug.log*
# env files (can opt-in for committing if needed)
.env*
# vercel
.vercel
# typescript
*.tsbuildinfo
next-env.d.ts

View File

@@ -1,36 +0,0 @@
This is a [Next.js](https://nextjs.org) project bootstrapped with [`create-next-app`](https://nextjs.org/docs/app/api-reference/cli/create-next-app).
## Getting Started
First, run the development server:
```bash
npm run dev
# or
yarn dev
# or
pnpm dev
# or
bun dev
```
Open [http://localhost:3000](http://localhost:3000) with your browser to see the result.
You can start editing the page by modifying `app/page.tsx`. The page auto-updates as you edit the file.
This project uses [`next/font`](https://nextjs.org/docs/app/building-your-application/optimizing/fonts) to automatically optimize and load [Geist](https://vercel.com/font), a new font family for Vercel.
## Learn More
To learn more about Next.js, take a look at the following resources:
- [Next.js Documentation](https://nextjs.org/docs) - learn about Next.js features and API.
- [Learn Next.js](https://nextjs.org/learn) - an interactive Next.js tutorial.
You can check out [the Next.js GitHub repository](https://github.com/vercel/next.js) - your feedback and contributions are welcome!
## Deploy on Vercel
The easiest way to deploy your Next.js app is to use the [Vercel Platform](https://vercel.com/new?utm_medium=default-template&filter=next.js&utm_source=create-next-app&utm_campaign=create-next-app-readme) from the creators of Next.js.
Check out our [Next.js deployment documentation](https://nextjs.org/docs/app/building-your-application/deploying) for more details.

View File

@@ -1,103 +0,0 @@
import Image from "next/image";
export default function Home() {
return (
<div className="font-sans grid grid-rows-[20px_1fr_20px] items-center justify-items-center min-h-screen p-8 pb-20 gap-16 sm:p-20">
<main className="flex flex-col gap-[32px] row-start-2 items-center sm:items-start">
<Image
className="dark:invert"
src="/next.svg"
alt="Next.js logo"
width={180}
height={38}
priority
/>
<ol className="font-mono list-inside list-decimal text-sm/6 text-center sm:text-left">
<li className="mb-2 tracking-[-.01em]">
Get started by editing{" "}
<code className="bg-black/[.05] dark:bg-white/[.06] font-mono font-semibold px-1 py-0.5 rounded">
app/page.tsx
</code>
.
</li>
<li className="tracking-[-.01em]">
Save and see your changes instantly.
</li>
</ol>
<div className="flex gap-4 items-center flex-col sm:flex-row">
<a
className="rounded-full border border-solid border-transparent transition-colors flex items-center justify-center bg-foreground text-background gap-2 hover:bg-[#383838] dark:hover:bg-[#ccc] font-medium text-sm sm:text-base h-10 sm:h-12 px-4 sm:px-5 sm:w-auto"
href="https://vercel.com/new?utm_source=create-next-app&utm_medium=appdir-template-tw&utm_campaign=create-next-app"
target="_blank"
rel="noopener noreferrer"
>
<Image
className="dark:invert"
src="/vercel.svg"
alt="Vercel logomark"
width={20}
height={20}
/>
Deploy now
</a>
<a
className="rounded-full border border-solid border-black/[.08] dark:border-white/[.145] transition-colors flex items-center justify-center hover:bg-[#f2f2f2] dark:hover:bg-[#1a1a1a] hover:border-transparent font-medium text-sm sm:text-base h-10 sm:h-12 px-4 sm:px-5 w-full sm:w-auto md:w-[158px]"
href="https://nextjs.org/docs?utm_source=create-next-app&utm_medium=appdir-template-tw&utm_campaign=create-next-app"
target="_blank"
rel="noopener noreferrer"
>
Read our docs
</a>
</div>
</main>
<footer className="row-start-3 flex gap-[24px] flex-wrap items-center justify-center">
<a
className="flex items-center gap-2 hover:underline hover:underline-offset-4"
href="https://nextjs.org/learn?utm_source=create-next-app&utm_medium=appdir-template-tw&utm_campaign=create-next-app"
target="_blank"
rel="noopener noreferrer"
>
<Image
aria-hidden
src="/file.svg"
alt="File icon"
width={16}
height={16}
/>
Learn
</a>
<a
className="flex items-center gap-2 hover:underline hover:underline-offset-4"
href="https://vercel.com/templates?framework=next.js&utm_source=create-next-app&utm_medium=appdir-template-tw&utm_campaign=create-next-app"
target="_blank"
rel="noopener noreferrer"
>
<Image
aria-hidden
src="/window.svg"
alt="Window icon"
width={16}
height={16}
/>
Examples
</a>
<a
className="flex items-center gap-2 hover:underline hover:underline-offset-4"
href="https://nextjs.org?utm_source=create-next-app&utm_medium=appdir-template-tw&utm_campaign=create-next-app"
target="_blank"
rel="noopener noreferrer"
>
<Image
aria-hidden
src="/globe.svg"
alt="Globe icon"
width={16}
height={16}
/>
Go to nextjs.org
</a>
</footer>
</div>
);
}

View File

@@ -1,7 +0,0 @@
import type { NextConfig } from "next";
const nextConfig: NextConfig = {
/* config options here */
};
export default nextConfig;

View File

@@ -1,32 +0,0 @@
{
"name": "project-noctivus",
"version": "0.1.0",
"private": true,
"scripts": {
"dev": "next dev --turbopack",
"build": "next build",
"start": "next start",
"lint": "next lint"
},
"dependencies": {
"class-variance-authority": "^0.7.1",
"clsx": "^2.1.1",
"lucide-react": "^0.539.0",
"next": "15.4.6",
"react": "19.1.0",
"react-dom": "19.1.0",
"tailwind-merge": "^3.3.1"
},
"devDependencies": {
"@eslint/eslintrc": "^3",
"@tailwindcss/postcss": "^4",
"@types/node": "^20",
"@types/react": "^19",
"@types/react-dom": "^19",
"eslint": "^9",
"eslint-config-next": "15.4.6",
"tailwindcss": "^4",
"tw-animate-css": "^1.3.6",
"typescript": "^5"
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,27 +0,0 @@
{
"compilerOptions": {
"target": "ES2017",
"lib": ["dom", "dom.iterable", "esnext"],
"allowJs": true,
"skipLibCheck": true,
"strict": true,
"noEmit": true,
"esModuleInterop": true,
"module": "esnext",
"moduleResolution": "bundler",
"resolveJsonModule": true,
"isolatedModules": true,
"jsx": "preserve",
"incremental": true,
"plugins": [
{
"name": "next"
}
],
"paths": {
"@/*": ["./*"]
}
},
"include": ["next-env.d.ts", "**/*.ts", "**/*.tsx", ".next/types/**/*.ts"],
"exclude": ["node_modules"]
}

View File

Before

Width:  |  Height:  |  Size: 391 B

After

Width:  |  Height:  |  Size: 391 B

View File

Before

Width:  |  Height:  |  Size: 1.0 KiB

After

Width:  |  Height:  |  Size: 1.0 KiB

View File

Before

Width:  |  Height:  |  Size: 1.3 KiB

After

Width:  |  Height:  |  Size: 1.3 KiB

View File

Before

Width:  |  Height:  |  Size: 128 B

After

Width:  |  Height:  |  Size: 128 B

View File

Before

Width:  |  Height:  |  Size: 385 B

After

Width:  |  Height:  |  Size: 385 B

View File

@@ -1,157 +0,0 @@
import * as dotenv from "dotenv";
dotenv.config();
import { loadStoryConfig } from "./config";
import { validatePublicDomain } from "./validator";
import { sanitizeText } from "./sanitizer";
import { chunkText } from "./chunker";
import { generateAudio, generateSingleAudio } from "./tts";
import { combineAudio, getChunkDuration } from "./audio";
import { generateImage } from "./images";
import { createVideo } from "./video";
import { createSrt } from "./subtitles";
import { generateYouTubeMetadata, uploadToYouTube } from "./uploader";
import * as path from "path";
import * as fs from "fs";
async function mapWithConcurrency<T>(
items: T[],
limit: number,
mapper: (item: T, index: number) => Promise<void>
): Promise<void> {
if (items.length === 0) return;
let nextIndex = 0;
const inFlight: Promise<void>[] = [];
const launchNext = () => {
if (nextIndex >= items.length) return;
const current = nextIndex++;
const p = mapper(items[current], current).finally(() => {
const idx = inFlight.indexOf(p);
if (idx >= 0) inFlight.splice(idx, 1);
});
inFlight.push(p);
};
for (let i = 0; i < Math.min(limit, items.length); i++) {
launchNext();
}
while (inFlight.length > 0 || nextIndex < items.length) {
while (inFlight.length < limit && nextIndex < items.length) {
launchNext();
}
await Promise.race(inFlight);
}
}
async function main() {
const rawArgs = process.argv.slice(2);
const storyName = rawArgs.find((a) => !a.startsWith("--"));
const force = rawArgs.includes("--force");
const skipUpload = rawArgs.includes("--skip-upload");
const concurrencyArg = rawArgs.find((a) => a.startsWith("--concurrency="));
const concurrency = concurrencyArg ? Math.max(1, parseInt(concurrencyArg.split("=")[1], 10) || 3) : 3;
if (!storyName) {
console.error("Usage: ts-node src/orchestrator.ts <storyName> [--force] [--skip-upload] [--concurrency=N]");
process.exit(1);
}
console.log(`Starting pipeline for story: ${storyName}`);
const storyConfig = loadStoryConfig(storyName);
console.log("Story configuration:");
console.log(storyConfig);
const validationResult = validatePublicDomain(storyConfig);
console.log(`Public domain status: ${validationResult.message}`);
if (!validationResult.is_public_domain) {
process.exit(1);
}
const storyRoot = path.resolve("stories", storyName);
["audio", "images", "final_audio", "video"].forEach((d) => fs.mkdirSync(path.join(storyRoot, d), { recursive: true }));
console.log("Sanitizing text...");
const sanitizedText = sanitizeText(storyName);
console.log("Sanitized text:");
console.log(sanitizedText);
console.log("Chunking text...");
const chunks = chunkText(sanitizedText, storyConfig.config.chunk_size);
console.log("Text chunks:");
console.log(chunks);
console.log("Generating intro/outro audio...");
const introFile = path.join("stories", storyName, storyConfig.config.intro_audio_file);
const outroFile = path.join("stories", storyName, storyConfig.config.outro_audio_file);
if (!fs.existsSync(introFile) || force) {
await generateSingleAudio(storyConfig, "This is the intro.", introFile);
} else {
console.log(`Skipping intro generation, exists: ${introFile}`);
}
if (!fs.existsSync(outroFile) || force) {
await generateSingleAudio(storyConfig, "This is the outro.", outroFile);
} else {
console.log(`Skipping outro generation, exists: ${outroFile}`);
}
console.log("Intro/outro audio ready.");
console.log(`Generating ${chunks.length} audio chunks with concurrency=${concurrency}...`);
const audioFiles: string[] = new Array(chunks.length);
const chunkDurations: number[] = new Array(chunks.length);
await mapWithConcurrency(chunks, concurrency, async (chunk, i) => {
const audioPath = path.join("stories", storyName, "audio", `chunk_${i}.mp3`);
if (!fs.existsSync(audioPath) || force) {
console.log(`Generating audio for chunk ${i}...`);
await generateAudio(storyConfig, storyName, chunk, i);
} else {
console.log(`Skipping audio for chunk ${i}, exists.`);
}
const duration = await getChunkDuration(audioPath);
audioFiles[i] = audioPath;
chunkDurations[i] = duration;
console.log(`Audio chunk ${i} ready: ${audioPath}, duration: ${duration}`);
});
console.log(`Generating ${chunks.length} images with concurrency=${concurrency}...`);
const imageFiles: string[] = new Array(chunks.length);
await mapWithConcurrency(chunks, concurrency, async (chunk, i) => {
const imagePath = path.join("stories", storyName, "images", `chunk_${i}_img0.png`);
if (!fs.existsSync(imagePath) || force) {
console.log(`Generating image for chunk ${i}...`);
const generated = await generateImage(storyName, storyConfig, chunk, i, 0);
imageFiles[i] = generated;
} else {
console.log(`Skipping image for chunk ${i}, exists.`);
imageFiles[i] = imagePath;
}
console.log(`Image ${i} ready: ${imageFiles[i]}`);
});
console.log("Creating subtitles...");
const srtPath = createSrt(storyName, chunks, chunkDurations);
console.log(`Created subtitles file: ${srtPath}`);
console.log("Combining audio files...");
await combineAudio(storyName, storyConfig, audioFiles);
console.log("Combined audio files successfully.");
console.log("Creating video...");
await createVideo(storyName, storyConfig, imageFiles, chunkDurations, srtPath);
console.log("Created video successfully.");
if (skipUpload) {
console.log("Skipping upload step (--skip-upload).");
return;
}
console.log("Generating YouTube metadata...");
const metadata = generateYouTubeMetadata(storyConfig);
console.log("YouTube metadata:");
console.log(metadata);
console.log("Uploading to YouTube...");
const videoPath = path.resolve("stories", storyName, "video", "final.mp4");
await uploadToYouTube(videoPath, metadata);
console.log("YouTube upload placeholder complete.");
}
main();

View File

@@ -1,12 +1,27 @@
{ {
"compilerOptions": { "compilerOptions": {
"target": "es6", "target": "ES2017",
"module": "commonjs", "lib": ["dom", "dom.iterable", "esnext"],
"strict": true, "allowJs": true,
"esModuleInterop": true,
"skipLibCheck": true, "skipLibCheck": true,
"forceConsistentCasingInFileNames": true, "strict": true,
"outDir": "./dist" "noEmit": true,
"esModuleInterop": true,
"module": "esnext",
"moduleResolution": "bundler",
"resolveJsonModule": true,
"isolatedModules": true,
"jsx": "preserve",
"incremental": true,
"plugins": [
{
"name": "next"
}
],
"paths": {
"@/*": ["./*"]
}
}, },
"include": ["src/**/*.ts"] "include": ["next-env.d.ts", "**/*.ts", "**/*.tsx", ".next/types/**/*.ts", "types/**/*.d.ts"],
"exclude": ["node_modules"]
} }

4
types/ffprobe-static.d.ts vendored Normal file
View File

@@ -0,0 +1,4 @@
declare module "ffprobe-static" {
const value: { path: string };
export default value;
}