소프트웨어 정보

LiveCap is a real-time subtitle generation tool designed for streamers and gamers! With powerful speech recognition engines and translation features, break down language barriers and communicate with the world!!

All processing runs locally on your PC. No monthly fees, ever.

What LiveCap Can Do

🎮 Real-time Subtitles

Just speak and subtitles appear in real-time. Watch text grow as you talk, then solidify when confirmed. Elevate your stream's engagement instantly!

🌍 Real-time Translation

Speak in your language, get subtitles in another! Supports 14 languages (Japanese, English, Chinese, Korean, German, French, Spanish, and more). Connect with international viewers effortlessly. Choose between Google Translate and NVIDIA Riva.

📹 OBS Integration

Simple browser source setup for seamless integration. Fully customizable fonts, colors, strokes, shadows, and transparency. WebSocket API for advanced tool integration.

🎭 VRChat Support

Display subtitles in Chatbox via OSC. Perfect for international events! Real-time translation lets you enjoy conversations without language barriers.

📝 Video File Subtitle Generation

Drag & drop videos to auto-generate SRT subtitle files. YouTube-ready format output. Save hours on clip editing!

🎙️ Online Meetings & Presentations

Never miss important details with real-time subtitles. Easy meeting minutes creation. Accessibility support for hearing-impaired participants.

Choose from 11 Speech Recognition Engines

Select the perfect engine for your needs and achieve optimal recognition accuracy:

Japanese Specialized

• NVIDIA Parakeet TDT CTC 0.6B JA - Trained on 35,000 hours of Japanese audio for supreme accuracy

• ReazonSpeech K2 v2 - Lightweight and fast, runs smoothly on low-spec PCs

English Specialized

• NVIDIA Parakeet TDT 0.6B v2 - Ultra-high accuracy with WER 6.05%, professional-grade transcription

Multilingual

• MistralAI Voxtral Mini 3B - Supports 8 languages with automatic language detection

• NVIDIA Canary 1B Flash - Fast processing for English, German, French, and Spanish

• WhisperS2T (5 sizes) - OpenAI technology supporting 13+ languages, choose size based on your needs

Not sure which engine? Auto mode has you covered! Automatically detects language and selects the optimal engine.

Why Choose LiveCap?

🔒 Complete Privacy with Local Processing

Your voice data is never sent externally (text is only sent when using translation). Stream content and meetings stay secure.

⚡ Ultra-Low Latency

Local processing means no network delays. Enjoy smooth, real-time subtitle display.

💰 Completely Free, No Limits

No monthly fees, no character limits, no time restrictions (full version). Buy once, use forever.

🎨 Full Customization Freedom

Fonts, colors, sizes, positions—all at your fingertips. Perfect match for your streaming style.

Super Simple Setup!

• Launch and select microphone - No complicated configuration needed

• Choose an engine - Auto mode optimizes automatically

• Set output destination - OBS or VRChat, your choice

That's it! Start using immediately!

System Requirements

Minimum Requirements

• Windows 10/11 (64bit)

• Intel Core i5 8th Gen / AMD Ryzen 5 3600

• 8GB RAM

• 10GB Storage

Recommended Requirements

• Intel Core i7 10th Gen / AMD Ryzen 7 5700

• 16GB RAM

• NVIDIA GPU (CUDA support for faster processing)

For NVIDIA Riva Translation

• NVIDIA GPU with 8GB+ VRAM required

Supported Formats

• Video: MP4, MKV, MOV, AVI, WebM, FLV

• Audio: WAV, MP3, M4A, AAC, OGG, FLAC, WMA

• Output: SRT subtitle files, WebSocket (for real-time streaming)

Before You Buy

Try the free trial version! The trial has a 3-minute limit but includes all features. Test it on your system before purchasing.

Support

Need help? Join our Discord community! Get answers to questions, request features, report bugs, and connect with our active community.

Documentation and support links available on the Steam page!

Break language barriers and connect with the world!

Start real-time subtitling with LiveCap today!!

AI 생성 콘텐츠 사용 공개

개발사는 게임에서 AI 생성 콘텐츠를 사용하는 방법을 다음과 같이 설명합니다.

Our software employs Whisper, one of the general-purpose speech recognition models, to transcribe your voice into text in real time, and it generates subtitles based on your speech. The subtitles provided are accurate transcriptions of your spoken words.
※Please note that the subtitle generation feature must not be used for purposes that infringe upon others' rights, such as copyright infringement.