From 00cff1eb7eb1400ebef66d37b0138f5837c348be Mon Sep 17 00:00:00 2001
From: latte <latte@hiddenden.cafe>
Date: Sun, 5 Apr 2026 15:21:26 +0200
Subject: [PATCH] first audio thing that i want to try

---
 .env.example                   |  13 ++
 README.md                      |  45 +++++++
 scripts/tts_generate.py        | 215 +++++++++++++++++++++++++++++++++
 src/content/config.ts          |   1 +
 src/pages/blog/[...slug].astro |  28 +++++
 5 files changed, 302 insertions(+)
 create mode 100644 scripts/tts_generate.py

diff --git a/.env.example b/.env.example
index ee072c2..03f8e53 100644
--- a/.env.example
+++ b/.env.example
@@ -15,6 +15,19 @@ DB_PATH=/data/guestbook.db
 HOST=0.0.0.0
 PORT=3000
 
+# --- TTS (scripts/tts_generate.py) ---
+# Provider: "google" (default, 1M chars/month free), "mistral", or "openai"
+TTS_PROVIDER=google
+
+# Google Cloud TTS — https://console.cloud.google.com/apis/credentials
+GOOGLE_API_KEY=
+
+# Mistral Voxtral TTS — https://console.mistral.ai/api-keys
+# MISTRAL_API_KEY=
+
+# OpenAI TTS-1 — https://platform.openai.com/api-keys
+# OPENAI_API_KEY=
+
 # --- Development overrides ---
 # For local dev (npm run dev), override with:
 # COOKIE_SECURE=false
diff --git a/README.md b/README.md
index 54473f2..bab98a2 100644
--- a/README.md
+++ b/README.md
@@ -78,6 +78,51 @@ docker login git.hiddenden.cafe
 docker push git.hiddenden.cafe/mats/cozy-den:latest
 ```
 
+## Blog Text-to-Speech
+
+Blog posts kunnen automatisch omgezet worden naar audio met `scripts/tts_generate.py`.
+
+### Setup
+
+1. Kopieer `.env.example` naar `.env` en vul je API key in
+2. Exporteer de variabelen:
+```bash
+export GOOGLE_API_KEY=jouw_sleutel_hier
+# of: export MISTRAL_API_KEY=... / export OPENAI_API_KEY=...
+```
+
+### Audio genereren
+
+```bash
+# Eén post — output: <slug>.mp3 in de huidige map
+python scripts/tts_generate.py src/content/blog/hello-world.md
+
+# Andere output directory
+python scripts/tts_generate.py src/content/blog/hello-world.md --output-dir public/audio
+
+# Andere provider
+TTS_PROVIDER=openai python scripts/tts_generate.py src/content/blog/hello-world.md
+```
+
+Beschikbare providers (via `TTS_PROVIDER`):
+
+| Provider | Env var | Prijs |
+|---|---|---|
+| `google` (standaard) | `GOOGLE_API_KEY` | 1M chars/maand gratis |
+| `mistral` | `MISTRAL_API_KEY` | ~$16/M chars |
+| `openai` | `OPENAI_API_KEY` | ~$15/M chars |
+
+### Audio tonen op een blogpost
+
+Voeg `audio: true` toe aan het frontmatter van de post. De `<audio>` player verschijnt automatisch en laadt van `https://audio.hiddenden.cafe/<slug>.mp3`.
+
+```yaml
+---
+title: "Mijn post"
+audio: true
+---
+```
+
 ## Customization
 
 The site is built to be easily customizable:
diff --git a/scripts/tts_generate.py b/scripts/tts_generate.py
new file mode 100644
index 0000000..2023375
--- /dev/null
+++ b/scripts/tts_generate.py
@@ -0,0 +1,215 @@
+#!/usr/bin/env python3
+"""
+Blog Text-to-Speech generator.
+
+Usage:
+    python scripts/tts_generate.py src/content/blog/my-post.md
+
+Environment variables:
+    TTS_PROVIDER   - "google" (default), "mistral", or "openai"
+    GOOGLE_API_KEY - Required when TTS_PROVIDER=google
+    MISTRAL_API_KEY - Required when TTS_PROVIDER=mistral
+    OPENAI_API_KEY  - Required when TTS_PROVIDER=openai
+
+Output:
+    <slug>.mp3 in the current working directory (or --output-dir if specified)
+"""
+
+import argparse
+import os
+import re
+import sys
+
+
+def parse_frontmatter(text: str) -> tuple[dict, str]:
+    """Extract YAML frontmatter and return (metadata_dict, body)."""
+    if not text.startswith("---"):
+        return {}, text
+
+    end = text.find("\n---", 3)
+    if end == -1:
+        return {}, text
+
+    front = text[3:end].strip()
+    body = text[end + 4:].strip()
+
+    meta: dict = {}
+    for line in front.splitlines():
+        if ":" in line:
+            key, _, value = line.partition(":")
+            meta[key.strip()] = value.strip().strip('"').strip("'")
+
+    return meta, body
+
+
+def clean_markdown(text: str) -> str:
+    """Strip markdown syntax so TTS reads clean prose."""
+    # Remove code blocks (``` ... ```)
+    text = re.sub(r"```[\s\S]*?```", "", text)
+    # Remove inline code
+    text = re.sub(r"`[^`]+`", "", text)
+    # Remove images
+    text = re.sub(r"!\[.*?\]\(.*?\)", "", text)
+    # Convert links to just the link text
+    text = re.sub(r"\[([^\]]+)\]\([^)]+\)", r"\1", text)
+    # Remove ATX headings markers but keep text
+    text = re.sub(r"^#{1,6}\s+", "", text, flags=re.MULTILINE)
+    # Remove bold/italic markers
+    text = re.sub(r"\*{1,3}([^*]+)\*{1,3}", r"\1", text)
+    text = re.sub(r"_{1,3}([^_]+)_{1,3}", r"\1", text)
+    # Remove horizontal rules
+    text = re.sub(r"^[-*_]{3,}\s*$", "", text, flags=re.MULTILINE)
+    # Remove blockquote markers
+    text = re.sub(r"^>\s?", "", text, flags=re.MULTILINE)
+    # Remove list markers
+    text = re.sub(r"^[\*\-\+]\s+", "", text, flags=re.MULTILINE)
+    text = re.sub(r"^\d+\.\s+", "", text, flags=re.MULTILINE)
+    # Collapse multiple blank lines
+    text = re.sub(r"\n{3,}", "\n\n", text)
+    return text.strip()
+
+
+def tts_google(text: str, slug: str, output_path: str) -> None:
+    """Generate audio with Google Cloud TTS (free tier: 1M chars/month)."""
+    import json
+    import urllib.request
+
+    api_key = os.environ.get("GOOGLE_API_KEY")
+    if not api_key:
+        raise EnvironmentError("GOOGLE_API_KEY is not set")
+
+    payload = json.dumps({
+        "input": {"text": text},
+        "voice": {
+            "languageCode": "nl-NL",
+            "name": "nl-NL-Wavenet-D",
+            "ssmlGender": "FEMALE",
+        },
+        "audioConfig": {"audioEncoding": "MP3"},
+    }).encode()
+
+    url = f"https://texttospeech.googleapis.com/v1/text:synthesize?key={api_key}"
+    req = urllib.request.Request(url, data=payload, headers={"Content-Type": "application/json"})
+
+    with urllib.request.urlopen(req) as resp:
+        body = json.loads(resp.read())
+
+    import base64
+    audio_bytes = base64.b64decode(body["audioContent"])
+    with open(output_path, "wb") as f:
+        f.write(audio_bytes)
+
+
+def tts_mistral(text: str, slug: str, output_path: str) -> None:
+    """Generate audio with Mistral Voxtral TTS (~$16/M chars)."""
+    import json
+    import urllib.request
+
+    api_key = os.environ.get("MISTRAL_API_KEY")
+    if not api_key:
+        raise EnvironmentError("MISTRAL_API_KEY is not set")
+
+    payload = json.dumps({
+        "model": "voxtral-mini-tts-2507",
+        "input": text,
+        "voice": "river",
+    }).encode()
+
+    url = "https://api.mistral.ai/v1/audio/speech"
+    req = urllib.request.Request(
+        url,
+        data=payload,
+        headers={
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {api_key}",
+        },
+    )
+
+    with urllib.request.urlopen(req) as resp:
+        audio_bytes = resp.read()
+
+    with open(output_path, "wb") as f:
+        f.write(audio_bytes)
+
+
+def tts_openai(text: str, slug: str, output_path: str) -> None:
+    """Generate audio with OpenAI TTS-1 (~$15/M chars)."""
+    import json
+    import urllib.request
+
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        raise EnvironmentError("OPENAI_API_KEY is not set")
+
+    payload = json.dumps({
+        "model": "tts-1",
+        "input": text,
+        "voice": "nova",
+    }).encode()
+
+    url = "https://api.openai.com/v1/audio/speech"
+    req = urllib.request.Request(
+        url,
+        data=payload,
+        headers={
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {api_key}",
+        },
+    )
+
+    with urllib.request.urlopen(req) as resp:
+        audio_bytes = resp.read()
+
+    with open(output_path, "wb") as f:
+        f.write(audio_bytes)
+
+
+PROVIDERS = {
+    "google": tts_google,
+    "mistral": tts_mistral,
+    "openai": tts_openai,
+}
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Generate TTS audio for a blog post")
+    parser.add_argument("file", help="Path to the .md blog post")
+    parser.add_argument("--output-dir", default=".", help="Directory to write the .mp3 (default: .)")
+    args = parser.parse_args()
+
+    md_path = args.file
+    if not os.path.isfile(md_path):
+        print(f"ERROR: file not found: {md_path}", file=sys.stderr)
+        sys.exit(1)
+
+    with open(md_path, encoding="utf-8") as f:
+        raw = f.read()
+
+    meta, body = parse_frontmatter(raw)
+
+    slug = meta.get("slug") or os.path.splitext(os.path.basename(md_path))[0]
+    title = meta.get("title", "")
+
+    # Prepend title so TTS reads it aloud
+    full_text = f"{title}.\n\n{clean_markdown(body)}" if title else clean_markdown(body)
+
+    provider_name = os.environ.get("TTS_PROVIDER", "google").lower()
+    if provider_name not in PROVIDERS:
+        print(f"ERROR: unknown TTS_PROVIDER '{provider_name}'. Choose from: {', '.join(PROVIDERS)}", file=sys.stderr)
+        sys.exit(1)
+
+    output_path = os.path.join(args.output_dir, f"{slug}.mp3")
+
+    print(f"Generating audio for '{slug}' using provider '{provider_name}' ({len(full_text)} chars)...")
+
+    try:
+        PROVIDERS[provider_name](full_text, slug, output_path)
+    except Exception as exc:
+        print(f"ERROR: TTS generation failed: {exc}", file=sys.stderr)
+        sys.exit(1)
+
+    print(f"Saved: {output_path}")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/src/content/config.ts b/src/content/config.ts
index 576e52f..aea1cb2 100644
--- a/src/content/config.ts
+++ b/src/content/config.ts
@@ -17,6 +17,7 @@ const blog = defineCollection({
             })
             .optional(),
         draft: z.boolean().optional().default(false),
+        audio: z.boolean().optional().default(false),
     }),
 });
 
diff --git a/src/pages/blog/[...slug].astro b/src/pages/blog/[...slug].astro
index 0ee1501..1d33dc5 100644
--- a/src/pages/blog/[...slug].astro
+++ b/src/pages/blog/[...slug].astro
@@ -116,6 +116,23 @@ const readingTime = getReadingTime(post.body);
                     )
                 }
                 <div class="divider">==============================</div>
+
+                {
+                    post.data.audio && (
+                        <div class="audio-player">
+                            <audio
+                                controls
+                                preload="none"
+                                aria-label={`Luister naar: ${post.data.title}`}
+                            >
+                                <source
+                                    src={`https://audio.hiddenden.cafe/${post.slug}.mp3`}
+                                    type="audio/mpeg"
+                                />
+                            </audio>
+                        </div>
+                    )
+                }
             </header>
 
             <article class="content fade-in">
@@ -345,6 +362,17 @@ const readingTime = getReadingTime(post.body);
         border-color: color-mix(in srgb, var(--color-accent) 45%, transparent);
     }
 
+    .audio-player {
+        margin-top: var(--space-md);
+    }
+
+    .audio-player audio {
+        width: 100%;
+        height: 36px;
+        accent-color: var(--color-accent);
+        border-radius: 4px;
+    }
+
     .content {
         color: var(--color-text);
         line-height: 1.8;