[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-content-creator-ai-studio-en":3,"seo:pack:content-creator-ai-studio:en":97},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":96},"content-creator-ai-studio","🎬","#F43F5E","new","New · this week","Content Creator's AI Studio","Ten picks for the YouTuber, podcaster, newsletter writer, and TikTok creator who wants AI as a production crew: ideation, script, voiceover (cloud + open-source), captions, thumbnails, B-roll generation, and the publishing platform that holds it all together.",[16,28,36,44,51,58,66,73,81,89],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4297,"721d23c5-ffea-448c-b2a6-67c905855aad","claude-code-agent-content-marketer-721d23c5","Claude Code Agent: Content Marketer","Use this agent when you need to develop comprehensive content strategies, create SEO-optimized marketing content, or execute multi-channel content campaigns to drive engagement...","TokRepo精选",46,0,"en","skill","Skill",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":34,"view_count":35,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2978,"ad11ab44-64e2-4ece-9d8a-2a072fda98e3","elevenlabs-voice-design-generate-voices-from-prompts","ElevenLabs Voice Design — Generate Voices from Prompts","ElevenLabs Voice Design generates new voices from text prompts. Describe age, accent, tone — get a voice you own and reuse via TTS API.","ElevenLabs",103,{"id":37,"uuid":38,"slug":39,"title":40,"description":41,"author_name":42,"view_count":43,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2462,"e7a8aaaf-453a-11f1-9bc6-00163e2b0d79","styletts-2-human-level-text-speech-via-style-diffusion-e7a8aaaf","StyleTTS 2 — Human-Level Text-to-Speech via Style Diffusion","A TTS system that achieves human-level speech synthesis through style diffusion and adversarial training with large speech language models. Fast inference with natural prosody.","Script Depot",108,{"id":45,"uuid":46,"slug":47,"title":48,"description":49,"author_name":42,"view_count":50,"vote_count":24,"lang_type":25,"type":26,"type_label":27},390,"e1fd7c46-bbda-4956-8649-9c3ed579ff25","whisper-cpp-local-speech-text-pure-c-c-e1fd7c46","whisper.cpp — Local Speech-to-Text in Pure C\u002FC++","High-performance port of OpenAI Whisper in C\u002FC++. No Python, no GPU required. Runs on CPU, Apple Silicon, CUDA, and even Raspberry Pi. Real-time transcription.",1602,{"id":52,"uuid":53,"slug":54,"title":55,"description":56,"author_name":42,"view_count":57,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2458,"7e2317bb-453a-11f1-9bc6-00163e2b0d79","cogvideo-text-image-video-generation-7e2317bb","CogVideo — Text and Image to Video Generation","An open-source video generation framework from Zhipu AI supporting text-to-video and image-to-video with CogVideoX models. Generates high-quality clips up to 6 seconds.",155,{"id":59,"uuid":60,"slug":61,"title":62,"description":63,"author_name":64,"view_count":65,"vote_count":24,"lang_type":25,"type":26,"type_label":27},776,"84500559-5ce6-41c7-ba22-9712153bb821","together-ai-image-generation-skill-claude-code-84500559","Together AI Image Generation Skill for Claude Code","Skill that teaches Claude Code Together AI's image generation API. Covers FLUX and Kontext models for text-to-image, image editing, and style transfer with correct parameters.","Together AI",125,{"id":67,"uuid":68,"slug":69,"title":70,"description":71,"author_name":42,"view_count":72,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2348,"044138c3-43a5-11f1-9bc6-00163e2b0d79","imagemagick-command-line-image-processing-200-formats-044138c3","ImageMagick — Command-Line Image Processing for 200+ Formats","ImageMagick is a free, open-source software suite for creating, editing, compositing, and converting images. It supports over 200 image formats including PNG, JPEG, TIFF, WebP, SVG, and PDF.",163,{"id":74,"uuid":75,"slug":76,"title":77,"description":78,"author_name":79,"view_count":80,"vote_count":24,"lang_type":25,"type":26,"type_label":27},101,"7775f06a-8adf-477a-91e9-85f51682cd10","remotion-captions-subtitles-ai-powered-video-subtitles-7775f06a","Remotion Captions & Subtitles — AI-Powered Video Subtitles","AI skill for generating and rendering captions in Remotion videos. Supports transcription, word-level timing, and styled subtitle export.","Skill Factory",197,{"id":82,"uuid":83,"slug":84,"title":85,"description":86,"author_name":87,"view_count":88,"vote_count":24,"lang_type":25,"type":26,"type_label":27},1328,"300e919c-381e-11f1-9bc6-00163e2b0d79","ghost-professional-publishing-platform-modern-journalism-300e919c","Ghost — Professional Publishing Platform for Modern Journalism","Ghost is an open-source publishing platform built for professional publishers. It bundles a blazing-fast Node.js CMS, Substack-style paid memberships, email newsletters, and SEO — everything a modern publication needs, self-hosted.","AI Open Source",194,{"id":90,"uuid":91,"slug":92,"title":93,"description":94,"author_name":42,"view_count":95,"vote_count":24,"lang_type":25,"type":26,"type_label":27},2434,"05ad6f38-44f8-11f1-9bc6-00163e2b0d79","yt-dlp-feature-rich-audio-video-downloader-05ad6f38","yt-dlp — Feature-Rich Audio & Video Downloader","yt-dlp is a feature-rich command-line tool for downloading audio and video from thousands of websites. A community-maintained fork of youtube-dl with active development, format selection, post-processing, and SponsorBlock integration.",106,"tokrepo install pack\u002Fcontent-creator-ai-studio",{"pageType":98,"pageKey":8,"locale":25,"title":99,"metaDescription":100,"h1":101,"tldr":102,"bodyMarkdown":103,"faq":104,"schema":120,"internalLinks":125,"citations":138,"wordCount":151,"generatedAt":152},"pack","Content Creator's AI Studio — 10 Tools for YouTube, Podcast, Newsletter","Content Marketer agent, ElevenLabs Voice Design, StyleTTS 2, whisper.cpp, CogVideo, Together AI image gen, ImageMagick, Remotion Captions, Ghost, yt-dlp — the 10-asset studio a solo creator uses for ideation, script, voice, video, thumbnails, and distribution. Install via TokRepo.","Content Creator's AI Studio — A Solo Creator's Production Crew","Ten picks in a deliberate pipeline: ideation and script first, then voiceover (cloud + open-source backup), captions, B-roll generation, thumbnails, and a publishing platform that doubles as your newsletter. Built for the YouTuber\u002Fpodcaster\u002Fnewsletter writer who can't hire an editor.","## What's in this pack\n\nThis is the stack a solo creator builds when they realise the part-time editor isn't coming back. Ten picks chosen to cover **every step of the content pipeline** — from \"what should I make this week\" to \"the email went out, the video uploaded, the thumbnail rendered.\" Each pick does one job in a real production flow.\n\nThe pack deliberately pairs a **cloud option and an open-source fallback** for the two stages where API bills explode: voiceover and image generation. You start on the cloud version while you're figuring out your format, then switch to the self-hosted one once your weekly output makes the API line on your statement uncomfortable.\n\nIt also assumes you're a **multi-platform creator** by default: the same script becomes a video, a podcast episode, a newsletter blurb, and three TikTok clips. The repurposing tools (yt-dlp + whisper.cpp + Ghost) exist so that one Tuesday script earns you five Tuesday content pieces.\n\n## Install in this order (ideation → script → produce → edit → distribute)\n\n1. **Claude Code Agent: Content Marketer** — start here, because nothing else matters if the topic is wrong. A Claude Code subagent that turns a one-line idea into outline, hook, beats, and platform-specific variants (long-form script + 60-second cut + tweet thread).\n2. **ElevenLabs Voice Design** — the cloud TTS that sounds least like a robot reading. Use Voice Design to mint your own consistent narrator from a prompt; reuse that voice ID forever so your channel has a recognisable sonic signature.\n3. **StyleTTS 2** — the open-source TTS you switch to when the ElevenLabs bill crosses your comfort line. Style-diffusion-based, near human quality, runs on a single consumer GPU. Use it as the backup or for B-track narration where the voice doesn't need to be \"the\" voice.\n4. **whisper.cpp** — local speech-to-text. The unglamorous tool that does the most work: transcribes your raw recordings for editing, generates captions, and feeds the repurposing pipeline (transcript → newsletter blurb → tweet thread). Runs offline so your unedited B-roll never leaves your machine.\n5. **CogVideo** — text-and-image-to-video for the B-roll you don't have footage for. Six-second clips are enough to cover \"shot of a person at a desk\" and \"city street\" filler. Doesn't replace real footage; it replaces stock-library subscriptions.\n6. **Together AI Image Generation** — the thumbnail and channel-art engine. Hosted Flux\u002FSD models with a clean API, fast iteration, and pricing that makes sense at indie volume. Generate 8 thumbnail variants in a minute, pick the best, ship.\n7. **ImageMagick** — the command-line image processor that handles every \"resize this thumbnail to 1280×720, 1080×1080, and 1920×1080\" task. One bash line per platform. This is the tool you'll quietly use the most.\n8. **Remotion Captions & Subtitles** — burn-in captions for short-form. TikTok \u002F Shorts \u002F Reels viewers watch with sound off; captions are not optional. Remotion's caption renderer styles them as code so every video has consistent typography.\n9. **Ghost** — the publishing platform that also runs your newsletter. One source of truth: long-form post + the email that goes to subscribers + the SEO-ready public page. Replaces \"WordPress + Mailchimp + Buffer\" with one self-hostable Node app.\n10. **yt-dlp** — the repurpose-pipeline cornerstone. Pull your own past episodes, your guest's old talks, conference recordings you want to clip — yt-dlp handles every platform with the same command. Feeds whisper.cpp for transcription and the Remotion pipeline for clipping.\n\n## How they fit together (ASCII content pipeline)\n\n```\n           ┌── Content Marketer Agent ──┐\n           │ (idea → outline → script)   │\n           └──────────────┬──────────────┘\n                          ▼\n          ┌── ElevenLabs Voice Design ──┐\n          │   or StyleTTS 2 (OSS)        │\n          │   (script → narration WAV)   │\n          └──────────────┬───────────────┘\n                          ▼\n     ┌──── whisper.cpp (transcribe narration) ────┐\n     │            ▼                                │\n     │     SRT + plain text                        │\n     │       │            │                        │\n     │       ▼            ▼                        │\n     │  Captions       Newsletter draft           │\n     │  (Remotion)     (Ghost)                    │\n     │       │            │                        │\n     │       ▼            ▼                        │\n     │   B-roll        Subscriber email            │\n     │   (CogVideo)    + public post URL          │\n     └────────┬───────────────────────────────────┘\n              ▼\n  ┌── Together AI image gen ──┐\n  │   (thumbnail variants)     │\n  │            │               │\n  │            ▼               │\n  │     ImageMagick            │\n  │  (resize 1280×720 \u002F        │\n  │   1080×1080 \u002F 1920×1080)   │\n  └────────────────────────────┘\n              │\n              ▼\n         yt-dlp (later)\n         pulls back the published\n         video → clip → repurpose\n```\n\nThe two critical joins are: **whisper.cpp → captions + newsletter** (one transcript powers two outputs) and **Together AI → ImageMagick** (one generated thumbnail becomes three platform-sized variants). Get those two joins right and your effort-per-piece-of-content drops by half.\n\n## Tradeoffs you'll hit\n\n- **DIY AI voice vs human voiceover** — AI voice has crossed the \"not embarrassing\" line in 2026. It has not crossed the \"sounds like a person who cares about this topic\" line. For your narrator persona on a flagship channel, record yourself. For B-roll narration, ad spots in your podcast, or platforms where you can't show your face, AI voice is fine and ships 10x faster. Start with ElevenLabs Voice Design while you figure out which is which.\n- **AI thumbnail vs human designer** — Together AI gets you to \"acceptable\" in 60 seconds. A real designer gets you to \"clickable\" in 4-8 hours. For weekly cadence on a growing channel, AI thumbnails plus a 5-minute human pass (text overlay, crop, contrast) beat outsourcing on speed. Once you hit 100K subscribers and a CTR point is worth real money, hire the designer.\n- **ElevenLabs cost vs StyleTTS 2 self-host** — ElevenLabs is $22-99\u002Fmo at indie volume and the voice quality is genuinely better. StyleTTS 2 is free after the GPU you already own and the voice quality is *almost* there. Rule of thumb: stay on ElevenLabs while you're under 30 minutes of generated audio\u002Fweek. Above that, the self-host math wins.\n- **Ghost vs Substack\u002FBeehiiv** — Ghost is the open-source self-hosted bet: you own the audience list and the platform never changes the rules on you. The cost: you maintain a server. Substack is the rent-the-audience bet: zero ops, but they can change the deal any Tuesday. For a creator who already manages a website, Ghost. For a creator who never wants to touch DNS, Substack.\n- **CogVideo vs licensed stock footage** — CogVideo clips are good for filler establishing shots. They're not good when the viewer is *looking at* the clip (close-ups, faces, specific actions). Budget $20-40\u002Fmo on a real stock library like Pexels Pro or Artgrid for the shots that carry meaning; use CogVideo for the cutaways.\n\n## Common pitfalls (sounding AI, license traps)\n\n- **Your videos start sounding AI.** Symptoms: every script uses \"dive into,\" \"in this video we'll explore,\" or \"the world of.\" Cause: the model's defaults leaking into your voice. Fix: maintain a `style.md` your Content Marketer agent reads first — explicit \"never use these phrases,\" \"sentence-cadence example,\" \"your hot takes are X, Y, Z.\" Update it monthly.\n- **License traps on generated voices\u002Fimages.** ElevenLabs voices generated by their Voice Design tool are yours to use commercially, but voices *cloned from another person* without consent are off-limits and platform-bannable. Same with Together AI image gen: most models allow commercial use, but the trained data is murky enough that you should not generate \"in the style of [living artist].\"\n- **Captions burnt in at the wrong aspect ratio.** Rendering 16:9 captions then cropping to 9:16 for Shorts cuts off the right half. Always render at the final platform aspect ratio; the Remotion Captions skill has explicit `width`\u002F`height` props — use them.\n- **Forgetting to re-encode for the platform.** YouTube wants H.264 + AAC at high bitrate; TikTok prefers a slightly lower bitrate and aggressive web-optimisation. Same source file, different exports. Add an ImageMagick-equivalent ffmpeg step per platform; don't upload one master MP4 and hope.\n- **Newsletter and video get out of sync.** You publish the video Tuesday, the newsletter goes out Friday referencing it, but you forgot to update the thumbnail in Ghost. Fix: one Ghost post per piece, and the email is generated *from* that post. Don't draft them in parallel.\n- **Repurposing without rewriting.** Pasting your video transcript verbatim into a newsletter is the laziest possible move and reads like one. Use the Content Marketer agent to *rewrite* the transcript into newsletter voice (shorter sentences, no \"as I said in the video,\" a fresh hook). Reuse the ideas, not the prose.",[105,108,111,114,117],{"q":106,"a":107},"Do I need all 10 tools or can I start smaller?","Start with four: Content Marketer agent for scripts, ElevenLabs Voice Design for narration, whisper.cpp for transcripts and captions, and Ghost for publishing + newsletter. That covers a working YouTube + newsletter pipeline. Add Together AI + ImageMagick when you get tired of making thumbnails by hand, add Remotion Captions when you start shipping Shorts\u002FReels, add CogVideo when you need B-roll, add StyleTTS 2 when the ElevenLabs bill starts to sting, and add yt-dlp the moment you decide to repurpose old episodes. The full 10 only makes sense once you're publishing more than one piece a week.",{"q":109,"a":110},"What does this actually cost per month for a solo creator?","Realistic baseline at one video + one newsletter per week: $5 Hetzner box for Ghost, $0 for whisper.cpp \u002F StyleTTS 2 \u002F ImageMagick \u002F yt-dlp (all open source), $22\u002Fmo for ElevenLabs starter, ~$5\u002Fmo for Together AI image gen at indie volume, and $0-10\u002Fmo for CogVideo API credits as needed. Call it $40\u002Fmo all-in, plus your Claude or GPT subscription for the Content Marketer agent. The cost line that grows fastest is ElevenLabs; that's the one StyleTTS 2 exists to replace.",{"q":112,"a":113},"Will the AI-generated voice get my YouTube channel demonetized?","Not on its own. YouTube's stance as of 2026: AI-generated content is fine for monetisation if it has clear creative input and isn't \"mass-produced, repetitive, or low-effort.\" An AI narrator on a script you wrote, in a video you edited, with your editorial point of view is not at risk. What does get demonetised: 20 channels uploading the same AI-generated script with the same AI-generated voice over the same AI-generated B-roll. Your taste is the moat.",{"q":115,"a":116},"Why Ghost instead of just Substack?","Substack is faster to start with — sign up, write, send. Ghost requires you to run a server (or pay Ghost Pro $9-25\u002Fmo). Why Ghost anyway: (1) you own the subscriber list outright with no platform between you and them, (2) it's a real CMS so the same post becomes your public page with proper SEO, not just an email archive, (3) the platform can't change its revenue share or content policy on you. If you plan to build for 5+ years, Ghost. If you want to publish next Tuesday, Substack.",{"q":118,"a":119},"Can the captions, voiceover, and B-roll all be generated from one script?","Yes — that's the whole point of wiring this pipeline. Your Content Marketer agent produces the script. The script goes to ElevenLabs (or StyleTTS 2) to produce the narration WAV. The WAV goes to whisper.cpp to produce a word-timed SRT (which becomes captions via Remotion). The script also goes to CogVideo as scene prompts for B-roll clips. One Tuesday script, automated narration, automated captions, automated B-roll suggestions. You're still in the loop for taste — picking the best take, fixing the B-roll that doesn't match — but the manual transcription\u002Ftiming\u002Fclip-hunting work is gone.",{"@context":121,"@type":122,"name":13,"description":123,"numberOfItems":124,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Ten AI assets curated for solo creators (YouTube \u002F podcast \u002F newsletter \u002F TikTok): ideation, script, voiceover, captions, thumbnails, B-roll, and publishing.",10,[126,130,134],{"url":127,"anchor":128,"reason":129},"\u002Fen\u002Fpacks\u002Fvideo-production-ai","Video Production AI pack","Once your pipeline outgrows hand-rendered videos, Remotion\u002FMoviePy\u002FOpenMontage cover the assembly-line side",{"url":131,"anchor":132,"reason":133},"\u002Fen\u002Fpacks\u002Fvoice-ai-stack","Voice AI Stack pack","Real-time voice agents (LiveKit, Cartesia, Moshi) extend the TTS picks here into live podcasting and voice chat",{"url":135,"anchor":136,"reason":137},"\u002Fen\u002Ffeatured","Featured assets on TokRepo","These ten picks live alongside the broader curated catalog of agent-ready creator tools",[139,143,147],{"claim":140,"source_name":141,"source_url":142},"ElevenLabs Voice Design generates a custom voice from a text prompt and persists a reusable voice ID","ElevenLabs Voice Design documentation","https:\u002F\u002Felevenlabs.io\u002Fdocs\u002Fproduct-guides\u002Fvoices\u002Fvoice-design",{"claim":144,"source_name":145,"source_url":146},"whisper.cpp is a high-performance, dependency-free C\u002FC++ port of OpenAI's Whisper that runs locally","ggerganov\u002Fwhisper.cpp on GitHub","https:\u002F\u002Fgithub.com\u002Fggerganov\u002Fwhisper.cpp",{"claim":148,"source_name":149,"source_url":150},"Ghost is an open-source publishing platform built for professional publishers with a built-in newsletter engine","Ghost official site","https:\u002F\u002Fghost.org\u002F",1380,"2026-05-22T07:30:00Z"]