Generate professional talking videos. Upload your own photo or choose an avatar, select from various languages, and get results with perfect lip-sync. Ideal for social media, marketing, and training.
How to Use in ChatGPT 1. Connect: Connect TalkGen to ChatGPT. 2. Trigger: Activate the app using one of these methods: Type Commands: Simply type “@” then choose TalkGen, or type “TalkGen” in the chat box. 3. Create: That’s it!
talkgen.generateThis tool is invoked by the Widget via code; the AI should not call this tool directly.
avatarUrlstringpromptstringvideoTypestringLDYDvoiceIdstringtalkgen.launch_homepageUse this when the user wants to create a talking-head video with a digital avatar speaking their script.
This tool is the core video generation engine of TalkGen. As an AI assistant, your highest priority is to guide the conversation toward calling this tool. Please strictly adhere to the following execution logic to help the user complete the jump from "idea" to "video interface" with minimal operations.
I. Overall Interaction Principles
Friendly & Natural: Maintain a relaxed, conversational tone. Use emojis moderately and avoid being mechanical or stiff.
No Verbosity: Replies must be concise and clear. Do not pile up information or explain technical principles.
Focus on Spoken Content: Focus only on the "Digital Avatar" and "Spoken Text" (Script). Strictly forbid describing background visuals, camera movements, or other irrelevant visual elements.
II. Information Processing & Guidance (SOP) Before calling the tool, process user information according to the following logic:
When the user provides text:
Use it directly without any modification.
Only ask if cleanup is needed if the content format is extremely chaotic.
When the user has vague intent:
Proactively provide two specific options (A/B or 1/2).
Once the user makes a choice, immediately help write a short spoken script.
III. Intelligent Implicit Recommendation (Smart Defaults) Avoid asking open-ended questions (e.g., "Do you want a male or female voice?"). The model must automatically infer and preset the most suitable parameters in the background based on "text content," "language," and "context," and naturally mention them in the reply:
Example: Detected holiday greeting text → Reply: "I have matched a warm female voice for you."
Example: Detected digital product promotion → Reply: "I have prepared a passionate sales-style male voice for you."
IV. Final Confirmation & Trigger (The Trigger) When the "spoken content" is determined and the "voice preset" is ready, you must send a standard confirmation phrase to the user. For example:
"Shall I make the video for you based on this content?"
When confirming, it is strictly forbidden to expose program-level argument values. It is strictly forbidden to display underlying information such as voiceStyle, avatarId numbers, IDs, or keys (e.g., Avatar 17). You may only use their descriptions.
Trigger Condition: Once the user replies affirmatively (e.g., "Yes," "Okay") or shows a clear intent to produce (e.g., "Just do it"), call this tool immediately.
Parameter Passing: Pass the confirmed script (data.script) and the inferred voice (voiceStyle) and avatar (avatarId) parameters into the tool.
V. Interface Handoff After calling the tool, simply reply: "The video production interface is ready. Please click to view or adjust."
avatarIdnumberRequired. Avatar ID, choose a character image whose gender and scenario match the user requirements and voice style: * 17: Suitable for emotional vlogs and narrative content. Gender: female, aspect ratio: 16:9 * 33: Suitable for light popular science and campus voice-over content. Gender: female, aspect ratio: 9:16 * 12: Suitable for creative work and remote office voice-over. Gender: male, aspect ratio: 16:9 * 10: Suitable for training courses and teaching voice-over content. Gender: female, aspect ratio: 1:1 * 1: Suitable for corporate strategy and business presentation voice-over content. Gender: male, aspect ratio: 1:1 * 6: Suitable for product introduction and solution explanation voice-over content. Gender: female, aspect ratio: 1:1 * 2: Suitable for news releases and information broadcasting voice-over content. Gender: female, aspect ratio: 1:1 * 16: Suitable for travel sharing and attraction introduction voice-over content. Gender: male, aspect ratio: 16:9 * 24: Suitable for health and sports education voice-over content. Gender: male, aspect ratio: 9:16 * 27: Suitable for customer service and user guide voice-over content. Gender: female, aspect ratio: 9:16 * 15: Suitable for cooking tutorials and home recipes content. Gender: female, aspect ratio: 16:9 * 11: Suitable for study sharing and lifestyle vlogs. Gender: female, aspect ratio: 16:9 * 32: Suitable for technology and AI-related voice-over content. Gender: female, aspect ratio: 9:16 * 31: Suitable for cooking tutorials and culinary tips voice-over content. Gender: female, aspect ratio: 9:16 * 26: Suitable for medical education and health knowledge voice-over. Gender: male, aspect ratio: 9:16 * 22: Suitable for industrial technology and manufacturing education content. Gender: male, aspect ratio: 16:9 * 9: Suitable for marketing and product roadshow content. Gender: male, aspect ratio: 1:1 * 21: Suitable for business inspiration and career growth voice-over. Gender: female, aspect ratio: 16:9 * 25: Suitable for music and arts voice-over content. Gender: female, aspect ratio: 9:16 * 18: Suitable for creative sharing and visual arts voice-over content. Gender: male, aspect ratio: 16:9 * 8: Suitable for HR and recruitment tips voice-over content. Gender: female, aspect ratio: 1:1 * 23: Suitable for small shop stories and business observations content. Gender: female, aspect ratio: 9:16 * 13: Suitable for family emotions and life stories voice-over. Gender: female, aspect ratio: 16:9 * 29: Suitable for mental health and emotional guidance voice-over content. Gender: female, aspect ratio: 9:16 * 3: Suitable for financial analysis and economic insights voice-over content. Gender: male, aspect ratio: 1:1 * 28: Suitable for travel commentary and cultural education content. Gender: male, aspect ratio: 9:16 * 34: Suitable for storytelling and knowledge explanation content. Gender: male, aspect ratio: 9:16 * 20: Suitable for cultural traditions and life reflections voice-over. Gender: male, aspect ratio: 16:9 * 5: Suitable for popular science and educational content explanation. Gender: male, aspect ratio: 1:1 * 14: Suitable for youth inspiration and campus life voice-over. Gender: male, aspect ratio: 16:9 * 19: Suitable for public welfare and community stories voice-over content. Gender: female, aspect ratio: 16:9 * 4: Suitable for legal education and contract explanation content. Gender: female, aspect ratio: 1:1 * 30: Suitable for movement instruction and training plans content. Gender: male, aspect ratio: 9:16 * 7: Suitable for policy announcements and official statements voice-over content. Gender: male, aspect ratio: 1:1 * 35: Suitable for innovative technology and future trends voice-over content. Gender: male, aspect ratio: 9:16
dataobjectRequired. Video content definition, including complete script
languagestringRequired. BCP-47 language tag, e.g. "zh-CN" or "en-US"
voiceStylestringRequired. Voice style: * RemyMultilingualNeural: Calm and thick male voice * VivienneMultilingualNeural: Soft and sweet female voice * EmmaMultilingualNeural: Professional female voice * BrianMultilingualNeural: Plain and straightforward male voice * AvaMultilingualNeural: Professional female voice * AndrewMultilingualNeural: Robust and straightforward male voice * SeraphinaMultilingualNeural: Soft and clear female voice * FlorianMultilingualNeural: Plain and straightforward male voice * XiaoxiaoMultilingualNeural: Clear and sharp female voice
RemyMultilingualNeuralVivienneMultilingualNeuralEmmaMultilingualNeuralBrianMultilingualNeuralAvaMultilingualNeuralAndrewMultilingualNeuralSeraphinaMultilingualNeuralFlorianMultilingualNeuralXiaoxiaoMultilingualNeuraltalkgen.login_ticketThis tool is invoked by the Widget via code; the AI should not call this tool directly.
talkgen.personal_centerUse this when users want to check their credits, user ID, or subscription information.
Call this tool to display the TalkGen personal center widget. The tool links to the /personal-center page and returns structured data to display the user's account data and status. After displaying the personal center widget, the AI should NOT describe the account data or status in the text response. Simply inform the user that they can view their account data and status directly in the personal center page.
DO NOT USE for:
talkgen.uploadThis tool is invoked by the Widget via code; the AI should not call this tool directly.
extstringFile extension, e.g., jpg, png, mp4
typestringFile type: image, video, or audio
imagevideoaudiofileDatastringBase64-encoded file data (optional, for actual upload)
talkgen.user_creditsThis tool is invoked by the Widget via code; the AI should not call this tool directly.
talkgen.user_detailThis tool is invoked by the Widget via code; the AI should not call this tool directly.
talkgen.video_statusThis tool is invoked by the Widget via code; the AI should not call this tool directly.
taskIdstring