Vibing
The Intelligent Voice Input Method
智能语音输入法
More than a voice input method, Vibing understands your intention, and types what you actually mean.
不只是语音输入法，Vibing 理解你的意图，为你打出真正想说的话。
Just Speak It. 开口即用
Hold key to talk, release to insert text. 按住说话，松开即输入文本。
Windows Ctrl + Win
~
1
2
3
4
5
6
7
8
9
0
-
=
Bksp
Tab
Q
W
E
R
T
Y
U
I
O
P
[
]
\
Caps
A
S
D
F
G
H
J
K
L
;
'
Enter
Shift
Z
X
C
V
B
N
M
,
.
/
Shift
Ctrl
Win
Alt
Alt
Win
Menu
Ctrl
Mac Right ⌥ Option
~
1
2
3
4
5
6
7
8
9
0
-
=
del
tab
Q
W
E
R
T
Y
U
I
O
P
[
]
\
caps
A
S
D
F
G
H
J
K
L
;
'
return
shift
Z
X
C
V
B
N
M
,
.
/
shift
fn
⌃
⌥
⌘
⌘
⌥
←
↑↓
→
|
Auto-pasted ✓已自动粘贴 ✓
Vibing Handles the Rest. 你只管说，剩下的交给 Vibing。
LLM-Powered Rewriting
What you said
Rewriting does three things, um, first it removes filler words like um and uh, second it handles auto-corrections — no, self-corrections, and uh, third it like auto-formats lists
What Vibing types
Rewriting does three things:
1. Remove filler words like um and uh
2. Handle self-corrections
3. Auto-format lists
1. Remove filler words like um and uh
2. Handle self-corrections
3. Auto-format lists
你说的
Vibing 能做三件事，呃，第一个是去掉嗯啊之类的填充词，第二个是处理口误自我更—不对，自我纠正，嗯第三个就是自动把列表格式化
Vibing 输出
Vibing 能做三件事：
1. 去除嗯、啊等填充词
2. 处理自我纠正
3. 自动格式化列表
1. 去除嗯、啊等填充词
2. 处理自我纠正
3. 自动格式化列表
Context-Aware Hotwords
Vibing
Vibing
Priyatham Reddy: "The new build of Kaleido is ready for testing."
You:
Other voice tools
Got it pretty thumb, I'll test the collide-o build this afternoon.
VibingGot it Priyatham, I'll test the Kaleido build this afternoon.
Priyatham Reddy: "Kaleido 的新版本已经可以测试了。"
你:
其他语音输入
收到普利亚森，我下午测一下卡雷多的新版本。
Vibing收到 Priyatham，我下午测一下 Kaleido 的新版本。
Automatically extracts names, project names, places, and other hotwords from your context — and recognizes them accurately.
从上下文中自动提取人名、项目名、地名等热词，并准确识别。
App-Aware Formatting
"Ugh this bug is driving me crazy, I've been debugging for two hours and it turns out it was just a typo in the config file, I can't believe it"
💬 Chat → Keep the vibe
Ugh this bug is driving me crazy, I've been debugging for two hours and it turns out it was just a typo in the config file, I can't believe it 😩
📄 Document → Just the facts
Spent two hours debugging; root cause was a typo in the config file.
"我靠这个bug搞了我两小时，结果就是配置文件里一个拼写错误，我真的服了"
💬 聊天 → 保留情绪
我靠这个bug搞了我两小时，结果就是配置文件里一个拼写错误，我真的服了😩
📄 文档 → 只留事实
排查两小时，根本原因是配置文件中的拼写错误。
Multilingual & Mixed Input
Vibing supports 50+ languages. No need to set the language beforehand — por ejemplo, ahora estoy hablando en español and Vibing handles it perfectly. You can even mix languages in a single sentence, which is essential for bilingual teams, cross-border business, multicultural content creation, and more.
Vibing 支持五十多种语言。不需要提前设置语言，for example, I am speaking English now and Vibing handles it well. 也可以在同一句话里夹杂多种语言，Vibing 都能准确识别，这对于 vibe coding、各个学科的 frontier research、创意自媒体等 scenario 都非常关键。
Long-Form Dictation
Voice input
The longer you speak, the better it gets. Uh, first, up to 10 minutes of continuous input. Second, think while talking — pause, reflect, and even... [30 seconds of thinking] ...and even change your, well, not just change your mind — take back what you said entirely, and Vibing will sort it out. Third, done thinking, done writing.
Output
The longer you speak, the better it gets.
1. Up to 10 minutes of continuous input.
2. Think while talking — pause, reflect, and even change your mind as you go.
3. Done thinking, done writing.
1. Up to 10 minutes of continuous input.
2. Think while talking — pause, reflect, and even change your mind as you go.
3. Done thinking, done writing.
语音输入
我觉得有几点是适合 long-form dictation 这个事儿的。一是 Vibing 支持十分钟的输入。二是用户可以边思考边讲，可以不说话，安静地思考，思考完再讲。也可以思考完之后推翻刚才所说的，最后 Vibing 会把……会统一最终结果。
Vibing 输出
Vibing 支持长达 10 分钟的连续输入。你可以边思考边讲，也可以暂停思考后继续。甚至可以推翻之前的内容，Vibing 会自动整合为最终结果。
Translation Mode
Voice input (Español)
El modo de traducción de Vibing admite más de 60 idiomas. Presiona Ctrl + Alt + Win, habla en tu idioma y la traducción se escribe directamente. Todas las funciones de transcripción se mantienen: reescritura, reconocimiento de contexto, palabras clave y más.
Output (English)
Vibing's translation mode supports over 60 languages. Press Ctrl + Alt + Win, speak in your language, and it types out the translation directly. All the transcription features are still there — rewriting, context awareness, hotwords, everything.
Español → English
语音输入（中文）
Vibing 的翻译模式支持六十多种语言，按 Ctrl + Alt + Win，说中文，直接打出翻译。而且转录的所有功能都保留——改写、上下文感知、热词，全都有。
Output (English)
Vibing's translation mode supports over 60 languages. Press Ctrl + Alt + Win, speak Chinese, and it types the translation directly. All transcription features are preserved — rewriting, context awareness, hotwords, and more.
中文 → English
FAQ
1. Strong models — VibeVoice-ASR for speech recognition, plus a multimodal model for intelligent rewriting.
2. Context-aware — reads your active app, input field, and surrounding text to deliver results that fit your situation. 1. 强大的模型——VibeVoice-ASR 负责语音识别，多模态模型负责智能改写。
2. 上下文感知——读取当前应用、输入框内容和周围文本，输出匹配你当前场景的结果。
2. Context-aware — reads your active app, input field, and surrounding text to deliver results that fit your situation. 1. 强大的模型——VibeVoice-ASR 负责语音识别，多模态模型负责智能改写。
2. 上下文感知——读取当前应用、输入框内容和周围文本，输出匹配你当前场景的结果。
Yes. Vibing works at the system level — any app that accepts text input, including editors, browsers, chat apps, terminals, and email.
是的。Vibing 在系统层面工作——任何接受文字输入的应用都支持，包括编辑器、浏览器、聊天软件、终端和邮件。
No. An internet connection is required for speech recognition and AI rewriting.
不可以。语音识别和 AI 改写需要联网。
不需要。Vibing 可以在中国大陆直接使用，无需 VPN。
50+ languages with automatic detection. Mixed-language input fully supported.
50+ 语言，自动检测语种，完全支持多语言混合输入。
All data is encrypted in transit. Audio and user context are deleted immediately after processing is complete. Nothing is stored or used for any other purpose.
所有数据在传输中均加密。音频和用户上下文在处理完成后立即删除，不会存储或用于任何其他用途。
Privacy隐私
Data processing:数据处理： To provide more accurate transcription, context-aware rewriting, and translation results, Vibing sends your audio and contextual information (such as screenshots, text in the active input field, and the current application name) to our servers. This data is used solely to process your request and return results. It is not retained after processing is complete.为了提供更准确的转录、上下文感知改写和翻译结果，Vibing 会将你的音频和上下文信息（如截图、当前输入框中的文本和当前应用名称）发送到我们的服务器。这些数据仅用于处理你的请求并返回结果，处理完成后不会保留。
Privacy commitment:隐私承诺： Your data is never stored or used for model training, analytics, or any other purpose beyond fulfilling your immediate request.你的数据不会被存储，也不会用于模型训练、数据分析或任何超出即时请求处理之外的用途。