闪电说 is an AI voice input app and communication Agent that lets you диктate on short press or draft replies with context on long press.
Shandianshuo is a communication Agent and also an AI voice input method. It puts "say it directly" and "help me say it" into the same workflow: a short press of the hotkey is used for voice input, while a long press is used to combine screen context, communication memory, and skills to organize a reply.
The product page emphasizes that it is not only about converting speech to text; it also organizes, corrects, and rewrites content based on the scenario so the result is clearer, more natural, and more appropriate. It is suitable for use in chat, email, documents, forms, customer service backends, and other input fields to help users complete everyday communication faster.
After short-pressing the hotkey and speaking, Shandianshuo converts your voice into text and types it into the current input field; the documentation says it performs light error correction, adds punctuation, and organizes the text, making it suitable for input scenarios where you already know what you want to say.
After long-pressing the hotkey and stating your task goal, Shandianshuo combines screen context and memory to generate a reply or rewrite content that is suitable for sending. It is designed for "help me say it" rather than literal transcription.
Before replying, you can load the other person's communication memory, such as work preferences, expression habits, key points, and historical context; after communication, these memories continue to update.
The knowledge base supports content such as style, dictionaries, and common replies: style controls the level of refinement and tone, dictionaries are used for names and product names, and common replies and materials can be added to the knowledge base.
Skills do more than rewrite text: they can also execute tasks and then organize a reply after completion, such as requesting data, running code, calling tools, or adopting someone's expression style.
It supports use in multiple input apps, and the page explicitly mentions scenarios such as WeChat, WeCom, DingTalk, Feishu, Qianniu, WhatsApp, browsers, email, DeepSeek, and terminals.
When you receive a message in WeChat, Feishu, or WeCom, long-press the hotkey and let Shandianshuo combine the chat context and memory to help you organize a reply that can be sent directly.
In email, comment sections, or customer service backends, first explain the task goal and tone requirements, then let Shandianshuo organize the content into a more complete and appropriate written expression.
In documents, web forms, search boxes, or terminals, short-press the hotkey and dictate the content directly so the text goes into the current input field quickly.
When names, product names, English abbreviations, or industry terms are easy to misrecognize, add the standard spelling to the dictionary to reduce repeated speaking and later corrections.
When a reply needs data lookup, code execution, or tool use first, use a skill so Shandianshuo completes the task first and then organizes the result into a sendable reply.
Shandianshuo supports short-pressing a hotkey to speak directly, and long-pressing a hotkey to let "help me say it" generate a reply you can send directly by combining screen context, memory, and skills. The getting-started page recommends first getting the model working, then testing the two core flows for short press and long press.
The official website says Shandianshuo can be used in chat, Feishu, DingTalk, WeCom, documents, email, comment sections, search boxes, customer service backends, web forms, and other text-input scenarios. The getting-started page also says it can work in various apps.
The official website says it supports macOS and Windows, the Android version is in beta, and the Linux version is under development. The download page also says the Mac version is available for Apple Silicon and Intel Macs.
The pricing page on the official website shows that Basic is free but requires manual setup; Pro works out of the box after logging in, with no key configuration needed; Max is marked as coming soon. Memberships, add-on packs, and redemption codes are all handled inside the product.
The official website says you can get the model working in three ways: a 3-day free membership, a subscription membership, or by configuring your own model. The getting-started page also mentions integrating speech recognition models, fast large models, and advanced large models, and supporting some third-party model services.
Twine AI Launcher is an Android launcher with AI assistant access, bringing ChatGPT-style help to your homescreen for quick answers, notes, tasks, reminders, and people memory.
Gemma AI is a phone call reminder app that calls you with scheduled reminders instead of push notifications. It helps people who want a more direct way to stay on schedule, with Google Calendar sync and conversational call interactions.
Nily AI is a browser-based AI assistant platform with 20+ models and tools for chat, writing, PDFs, email, images, and web page analysis. 7-day free trial.
小艺 is Huawei’s AI smart assistant for Q&A, writing, document reading, code help, image recognition, and file drag-and-drop.
MailerLite is an email marketing platform for sending campaigns, building automations, and creating landing pages, forms, and websites. It also supports digital products, paid newsletters, and recurring subscriptions.
AI Magicx is a unified AI workspace for chat, image, video, voice, music, email and developer tasks, helping teams and creators manage multiple models in one place.
