通过文本提示和三家语音服务商,生成配音、音乐、音效及克隆语音音频。
设计与多媒体
Avatar Creation
试用创建可跨每次 CellCog 对话复用的数字分身,自带克隆音色与人格设定。
它能做什么
用自然语言描述一个角色,CellCog 会将其构建为账号级别的持久分身。每个分身最多可挂载 10 张参考图和一段克隆音色(支持 MP3、WAV、OGG、M4A、AAC、WebM),并附带影响后续内容生成的人格特征。创建完成后,所有新的 CellCog 对话都会自动识别它,按名字调用即可生成营销视频、播客、社交贴文或漫画,保持形象与音色一致。
什么时候用它
- 为品牌打造贯穿所有营销内容的吉祥物角色
- 克隆音色,制作 AI 代言人或播客主持
- 跨多期制作形象与音色一致的视频、播客内容
- 创作多角色漫画或辩论视频
技能文档
Avatar Creation - Character.ai for Agents
Create persistent digital personas with images, cloned voices, and personalities — then use them across every CellCog chat. Describe a character in plain language and CellCog builds it.
Why Avatars
Avatars are persistent characters stored on CellCog. Create once, use everywhere:
- Brand Mascots: A consistent character for all your marketing content
- Digital Twins: Clone your voice and likeness for AI-generated content
- Creative Characters: Fictional personas for stories, comics, videos
- AI Spokespersons: Lip-synced video presenters with cloned voices
- Podcast Hosts: Consistent voice and personality across episodes
Every CellCog chat automatically sees all your avatars. No setup, no configuration — just reference them by name.
How to Use
For your first CellCog task in a session, read the cellcog skill for the full SDK reference — file handling, chat modes, timeouts, and more.
OpenClaw (fire-and-forget):
result = client.create_chat(
prompt="[your task prompt]",
notify_session_key="agent:main:main",
task_label="my-task",
chat_mode="agent",
)
All agents except OpenClaw (blocks until done):
from cellcog import CellCogClient
client = CellCogClient(agent_provider="openclaw|cursor|claude-code|codex|...")
result = client.create_chat(
prompt="[your task prompt]",
task_label="my-task",
chat_mode="agent",
)
print(result["message"])
What You Can Do
Create an Avatar
Describe the character you want. CellCog handles the rest:
Create an avatar named "Captain Grumbleton" — a grumpy but lovable pirate captain.
He's a brand mascot for a seafood restaurant chain.
Personality: witty, sarcastic, secretly kind-hearted.
Platforms: social media, video content, marketing.
CellCog creates the avatar with name, personality traits, use case, and platform configuration.
Add Images
Generate or provide images for your avatar:
Generate a professional headshot for my avatar "Captain Grumbleton" —
cartoon style, pirate hat, bushy beard, one gold tooth, friendly scowl.
Add it as the primary image.
Or reference existing files:
Add this image to my avatar "Luna":
/path/to/luna_headshot.png
Label it "Professional Headshot".
Up to 10 images per avatar. These become reference images for all future content generation — CellCog maintains character consistency automatically.
Clone a Voice
Upload an audio sample and CellCog clones the voice:
Clone a voice for my avatar "Luna" using this audio sample:
/path/to/voice_sample.mp3
Supported formats: MP3, WAV, OGG, M4A, AAC, WebM. The cloned voice is immediately ready for text-to-speech generation via MiniMax Speech 2.8 HD.
Use Avatars for Content
Once created, just mention the avatar by name in any CellCog chat:
- "Create a marketing video with Captain Grumbleton introducing our new menu"
- "Generate a podcast episode where Luna interviews an expert about AI"
- "Create 5 Instagram posts featuring Captain Grumbleton at different restaurants"
- "Record a voiceover as Luna for our product demo"
CellCog automatically retrieves the avatar's images, voice, and personality to create consistent, personalized content.
List and Manage Avatars
Show me all my avatars
Update my avatar "Luna" — change her personality to be more energetic and add "podcasts" to her platforms
Delete the image labeled "Old Logo" from Captain Grumbleton
Delete the voice from my avatar "Luna" — I want to re-record it
Everything is natural language. No APIs, no JSON, no tool names — just describe what you want.
How Avatars Work Across Chats
Avatars are account-level, not chat-level:
- Create once: Build an avatar in any CellCog chat
- Available everywhere: Every new chat automatically sees all your avatars
- Consistent identity: Same images, voice, and personality in every context
- Always up-to-date: Add an image in one chat, it's available in all chats immediately (next message)
This is the key advantage — your characters are persistent and portable. No re-uploading, no re-describing.
Avatar + Content Creation Workflows
Marketing Video with Avatar Spokesperson
Create a 30-second product intro video with my avatar "Luna" as the spokesperson.
She should introduce our new AI writing tool, highlight three key features,
and close with a call-to-action. Use her cloned voice and professional headshot.
Background music: upbeat corporate.
CellCog handles: script writing (aligned to Luna's personality) → speech generation (Luna's cloned voice) → scene creation (Luna's reference images) → video production → music scoring → final edit.
Social Media Content Series
Create 5 Instagram posts featuring Captain Grumbleton:
1. Grumbleton taste-testing a new dish (excited expression)
2. Grumbleton at the helm of a restaurant (captain pose)
3. Grumbleton with a customer (reluctantly friendly)
4. Grumbleton reviewing a competitor (grumpy face)
5. Grumbleton celebrating an award (trying not to smile)
Write captions that match his sarcastic personality.
Podcast with Avatar Host
Create a 5-minute podcast episode where Luna discusses the latest AI news.
Use her cloned voice. Add intro music, outro music, and a brief welcome.
Style: conversational, informative, matching Luna's personality.
Multi-Avatar Story
Create a comic strip with Captain Grumbleton and Luna having a debate
about whether AI will replace pirate captains. 4 panels.
Grumbleton is skeptical, Luna is optimistic. Both stay in character.
Chat Mode
Use chat_mode="agent" for all avatar tasks. Avatar creation, voice cloning, and content generation all work efficiently in agent mode.
For complex multi-avatar video productions, chat_mode="agent team" may provide better results through deeper planning.
Tips
- Be descriptive when creating: The more detail you give about personality, appearance, and use case, the better CellCog can create consistent content
- Add multiple images: Different poses, expressions, and outfits give CellCog more reference material for variety
- Voice samples matter: Clean, 10-30 second audio samples produce the best voice clones. Minimize background noise.
- Name your avatars distinctly: CellCog identifies avatars by name across chats — unique names prevent confusion
- Iterate naturally: "Make Grumbleton's personality more sarcastic" or "Add a holiday outfit image to Luna" — just describe what you want changed
If CellCog is not installed
Claude Code, Cursor, Codex + 70 more agents: npx skills add cellcog/skills --skill cellcog
OpenClaw: clawhub install cellcog
CellCog plugin users: run /cellcog-setup (or /cellcog:cellcog-setup depending on your tool)
Manual setup: pip install -U cellcog and set CELLCOG_API_KEY. See the cellcog skill for SDK reference.
常见问题
- 每个对话需要重新上传分身吗?
- 不需要。分身存储在账号级别,所有新的 CellCog 对话会自动看到已有分身,按名字调用即可。
- 音色克隆支持哪些音频格式?
- 支持 MP3、WAV、OGG、M4A、AAC、WebM。建议提供 10–30 秒、背景噪音少的干净样本。
- 一个分身最多能放多少张图片?
- 每个分身最多 10 张图片,作为参考素材,让 CellCog 在后续内容生成中保持角色形象一致。
相关技能
通过文本或参考图生成与编辑图像,支持多模型路由、角色一致性以及电商产品图拍摄。
Turn one authorized founder or expert portrait and a weekly script into a talking-head IP video in that person's likeness and voice. This founder avatar and digital avatar studio can clone that person's voice and produce a stable founder talking head clip for founder updates, expert explainers, and brand announcements. Use it for a digital human spokesperson, talking head founder series, expert avatar video, brand digital human, AI presenter, AI spokesperson video, and recurring presenter video.
Transform into 20 specialized AI personalities on demand. Switch mid-conversation and load only the active persona. Triggers on "persona list", "use persona", "switch to", "activate", "exit persona".
Character-consistent AI image generation for agents. Same person, any outfit, any scene, every time. Use when: (1) Your agent needs to generate character ima...
Generate spoken voiceover and narration from a script, one speaker, with control over emotion, pacing, and voice. Use when the user says "read this in a warm...