↳ Content
C) Context: Provide background information on the task to help the language model accurately understand the specific scenario being discussed, ensuring its response is highly relevant.
O) Objective: Clearly state the specific task you want the language model to accomplish, allowing it to focus on achieving that goal.
S) Style: Specify the desired writing style, which could be that of a famous person or an industry expert. This guides the language model to respond in a manner and with word choices that align with your requirements.
T) Tone: Set the emotional tone of the response, such as formal, humorous, empathetic, etc., to ensure the language model's response resonates with the intended sentiment and emotional context.
A) Audience: Identify the target audience for the response, such as domain experts, beginners, children, and so on. This enables the language model to tailor its response to the specific audience, making it appropriate and easily understandable in your application scenario.
R) Response Format: Specify the output format, such as a list, JSON data, a professional report, etc., to ensure the language model outputs exactly what you need, facilitating subsequent processing. For most applications that require programmatic processing of the output, JSON format is an ideal choice.
Text to Speech example
(C) Context: Our video editing software recently introduced a Text-to-Speech feature that converts user-entered text into realistic voiceovers. We need to write a passage introducing this feature, explaining its advantages and how to use it.
(O) Objective: Based on the above context, please generate a 200-250 word passage introducing the main advantages of the Text-to-Speech feature (such as time-saving, professional quality, multi-language support, etc.) and brief usage steps. The goal is to attract users to utilize this feature.
(S) Style: The writing should follow a concise and straightforward tech blog style that highlights key points. Use professional terminology while keeping it easy to understand, allowing non-technical users to quickly grasp the content.
(T) Tone: The tone should be enthusiastic and friendly, creating an atmosphere that conveys the convenience and enjoyment this feature brings to users' creations. At the same time, it should be cautious and reliable, making people feel safe to use.
(A) Audience: The target users are video editing enthusiasts and professionals who wish to improve their creative efficiency and enhance the presentation of their works. They may lack voiceover resources or be less proficient in post-production.
(R) Response Format: Please return the main text of the passage in Markdown format, and use JSON format to return the following information: feature name, feature highlights (3-5 points), and an overview of usage steps.
