Update to CellCog

{
  "description": "AI audio generation and text-to-speech powered by CellCog. Voiceover, narration, voice cloning, avatar voices, sound effects, music, podcasts, dialogue. Three voice providers (OpenAI, ElevenLabs, MiniMax). Professional audio production from text prompts.",
  "included_files": [],
  "name": "audio-generation-cellcog",
  "skill_md_contents": "---\nname: audio-generation-cellcog\ndescription: \"AI audio generation and text-to-speech powered by CellCog. Voiceover, narration, voice cloning, avatar voices, sound effects, music, podcasts, dialogue. Three voice providers (OpenAI, ElevenLabs, MiniMax). Professional audio production from text prompts.\"\n---\n# Audio Generation - AI Speech, Voiceover, and Sound Powered by CellCog\n\nCreate professional audio with AI — voiceovers, music, sound effects, and personalized avatar voices.\n\nCellCog provides **three voice providers**, each with different strengths. Choose based on your needs:\n\n| Scenario | Provider | Why |\n|----------|----------|-----|\n| Standard narration/voiceover | OpenAI | Best voice style control, consistent quality |\n| Emotional/dramatic delivery | ElevenLabs | Richest emotional range, supports emotion tags |\n| Cloned voice (avatar) | MiniMax | Only provider with voice cloning support |\n| Character voice with specific accent | ElevenLabs | 100+ diverse pre-made voices |\n| Fine pitch/speed/volume control | MiniMax | Granular voice settings |\n\n## How to Use\n\nUse the CellCog tools from this plugin's connector (`https://cellcog.ai/mcp`). The first time a tool runs, the user signs in with their CellCog account and clicks Approve.\n\n1. `cellcog_chat_create` with the brief as `prompt` (optionally `chat_mode` and `chat_tier`). It returns a chat id at once; the work runs in the background.\n2. Poll `cellcog_get_result` with that `chat_id` about once a minute. While it runs you get progress notes; when `complete` is true you get the final message and file links (valid 30 days), with previews of images and PDF pages.\n3. To iterate, `cellcog_chat_send` into the same chat. For files on the user's computer, open `cellcog_upload` and pass `upload:<upload_id>` in `attachments`; public https links can go in `attachments` directly.\n\nFor files, modes and tiers in depth, read the **cellcog** skill.\n\n---\n\n## Voice Providers\n\n### OpenAI (Default)\n\nBest for standard narration, voiceovers, and single-speaker content with precise delivery control.\n\n**Key strength**: Natural-language style instructions — describe the accent, tone, pacing, and emotion you want.\n\n**8 built-in voices:**\n\n| Voice | Gender | Characteristics |\n|-------|--------|----------------|\n| **cedar** | Male | Warm, resonant, authoritative, trustworthy |\n| **marin** | Female | Bright, articulate, emotionally agile, professional |\n| **ballad** | Male | Smooth, melodic, musical quality |\n| **coral** | Female | Vibrant, lively, dynamic, spirited |\n| **echo** | Male | Calm, measured, thoughtful, deliberate |\n| **sage** | Female | Wise, contemplative, reflective |\n| **shimmer** | Female | Soft, gentle, soothing, approachable |\n| **verse** | Male | Poetic, rhythmic, artistic, expressive |\n\nBest quality: **cedar** (male), **marin** (female).\n\n**Style customization examples:**\n- \"Warm conversational tone, medium pace, slight enthusiasm when mentioning features. American accent.\"\n- \"Deep, hushed, enigmatic, with a slow deliberate cadence — true crime narrator style.\"\n- \"Heavy French accent, sophisticated yet friendly, moderate pacing with deliberate pauses.\"\n\n---\n\n### ElevenLabs\n\nBest for emotional delivery, dramatic content, character voices, and audiobook narration.\n\n**Key strength**: Emotion tags embedded directly in text — `[laughs]`, `[sighs]`, `[whispers]`, `[excited]`, `[sarcastic]`. Plus 100+ diverse pre-made voices.\n\n**Emotion tags (use sparingly — 1-2 per paragraph):**\n\n| Tag | Effect |\n|-----|--------|\n| `[laughs]` | Natural laughter |\n| `[chuckles]` | Soft/brief laughter |\n| `[sighs]` | Sighing |\n| `[gasps]` | Surprise/shock |\n| `[whispers]` | Whispering delivery |\n| `[pause]` | Natural pause/beat |\n| `[sad]`, `[happy]`, `[excited]`, `[angry]`, `[sarcastic]` | Emotional delivery |\n\n**Example prompt:**\n> \"Generate speech using ElevenLabs with a warm British male voice:\n> 'And then, just when everyone thought it was over... [pause] [whispers] it wasn't.'\"\n\n---\n\n### MiniMax\n\nBest for cloned voices (avatars) and fine-grained voice control.\n\n**Key strength**: MiniMax Speech 2.8 HD — studio-grade audio quality. Supports avatar cloned voice IDs for personalized content, plus 17+ standard pre-made voices with granular speed, pitch, and volume control.\n\n**Standard voices include:** `Deep_Voice_Man`, `Calm_Woman`, `Casual_Guy`, `Lively_Girl`, `Wise_Woman`, `Friendly_Person`, `Young_Knight`, `Elegant_Man`, and more.\n\n**Voice settings:** emotion (happy/sad/angry/neutral/etc.), speed (0.5–2.0), volume (0–10), pitch (-12 to 12).\n\n---\n\n## Avatar / Cloned Voice\n\nUsers can create avatars on CellCog with their own cloned voice. When an avatar has a cloned voice, CellCog uses the MiniMax provider to generate speech that sounds like that person.\n\n**How it works:**\n- The user creates an avatar on cellcog.ai and uploads voice samples\n- CellCog clones their voice using MiniMax Speech 2.8 HD\n- Any audio request referencing that avatar uses their cloned voice\n\n**Example prompt:**\n> \"Generate a voiceover using my avatar Luna's voice: 'Welcome to our quarterly update. I'm excited to share some incredible results with you today.'\"\n\nThis is powerful for creating consistent, personalized content — marketing videos, podcast intros, course narration — all in the user's own voice.\n\n---\n\n## Sound Effects (SFX)\n\nCellCog generates standalone sound effects from text descriptions. Royalty-free, 0.1 to 30 seconds.\n\n**Example prompts:**\n- \"Generate a sound effect of heavy rain hitting a metal roof with occasional thunder, 10 seconds\"\n- \"Create a crispy footsteps-on-fresh-snow sound effect, 5 seconds\"\n- \"Generate an echoing door slam in a large empty warehouse\"\n\n**Tips for better SFX:**\n- Be specific about textures and environment\n- Specify duration when exact length matters\n- For ambient audio longer than 30 seconds, generate a short loopable segment and extend with ffmpeg\n\n---\n\n## Music Generation\n\nCreate original music from text descriptions. 3 seconds to 10 minutes. Royalty-free.\n\n**Capabilities:**\n- Any genre or genre fusion\n- Instrumental and vocal tracks (specify if you want vocals)\n- Complex arrangements, mood transitions, and energy dynamics\n- Describe what you want — the model handles music theory\n\n**Example prompts:**\n- \"Create 2 minutes of calm lo-fi hip-hop background music with soft piano and mellow beats, 75 BPM\"\n- \"Generate a 15-second upbeat tech podcast intro jingle\"\n- \"Create 90 seconds of cinematic orchestral music — start soft and inspiring, build to a confident crescendo\"\n- \"Generate a 3-minute pop song about summer adventures with female vocals\"\n\nFor precise section-by-section control (exact timing per section), describe your composition plan in detail — CellCog handles the structure.\n\n**All generated music is royalty-free** — use commercially without attribution or licensing fees.\n\n---\n\n## Multi-Language Support\n\nAll three voice providers support 40+ languages. Provide speech text in the target language:\n\nEnglish, Spanish, French, German, Italian, Portuguese, Chinese (Mandarin/Cantonese), Japanese, Korean, Hindi, Arabic, Russian, Polish, Dutch, Turkish, and many more.\n\n---\n\n## Choosing Mode & Tier\n\n| Scenario | Recommended |\n|----------|-------------|\n| Speech, narration, sound effects | `chat_mode=\"agent\"` (defaults to the `\"core\"` tier) |\n| Long-form productions with many segments | `chat_mode=\"agent\", chat_tier=\"max\"` |\n\n**The core tier is the default and handles most audio generation well**, balanced speed and depth. **Upgrade to `chat_tier=\"max\"`** when quality disappoints or the work is genuinely complex: just re-run the same prompt with `chat_tier=\"max\"`.\n\nAgent Team (`chat_mode=\"team\"`) is reserved for deep research — don't use it for audio generation.\n\n---\n\n## Tips for Better Audio\n\n1. **Choose the right provider**: OpenAI for standard narration, ElevenLabs for emotional/dramatic, MiniMax for cloned voices\n2. **Provide the complete script**: Write out exactly what should be spoken — don't say \"something about our product\"\n3. **Include style instructions**: \"Confident but warm\", \"slow and deliberate\", \"with slight excitement\"\n4. **For music**: Specify duration, mood, genre, and tempo (BPM if you know it)\n5. **Pronunciation guidance**: For names or technical terms, add hints: \"CellCog (pronounced SELL-kog)\"\n6. **For ElevenLabs emotion tags**: Use sparingly — 1-2 per paragraph. Tags affect all subsequent text until a new tag.\n"
}