Modern AI models now ingest staggering amounts of data at once.
graphic_eqSynthesizeeditEditdeleteDelete
Core Capabilities
Master the Kinetic Flow of Audio.
From ultra-precise transcription models to live translation, explore the engine powering the next generation of sound intelligence.
Real-time Transcription & Translation
Capture system audio and transform it into text instantly. Scribis provides millisecond-level transcription and renders translations simultaneously, keeping you in sync with the world.
TranscriptionLive speech recognition
Speaker 01你好，我正在使用 Scribis 的即時轉錄功能。
這項技術能瞬間捕捉每一句話。
並為您提供精準的語音識別結果。
Simultaneous TranslationEnglish → 繁體中文
Synced"Hi, I am using Scribis' real-time transcription feature."
你好，我正在使用 Scribis 的即時轉錄功能。
File Upload & Recording
Upload multiple media formats or record directly on our platform. From meeting notes to long-form podcasts, get high-precision results in minutes.
cloud_upload
Select Audio/Video
Drag & drop or click to upload (MP3, WAV, M4A, MP4)
interview_sample.wav78%
subtitles
Select Subtitles
Optional subtitle file (.srt, .vtt)
Optional ReferencelanguageLanguage
tuneModel Config
GPU ReadytuneAdvanced SettingsVRAM: ~4.5 GB
Voice Dictation & AI Assistant
Open voice dictation and the AI assistant with a shortcut. Turn spoken thoughts into text, then use voice commands to translate, rewrite, or summarize.
- translateTranslation AssistantSelect an email and say 'Translate this to Chinese' for instant, accurate results.
- edit_noteSmart EnhanceCorrect grammar, spelling, and remove fillers for professional content.
- history_eduCross-App UtilitySelect text in any application, anywhere, and invoke AI with your voice.
auto_fix_high
Before
Um... check the API spec for me, first step is to add the. underscore. user ID field, second step is to make a. correction. rule. colon. name cannot be empty. third step is to output the log. to. double dash. temp directory. also for priority. circle one. high. circle two. medium. that's basically it! Thanks!
south
Enhanced
Check the API spec for me. First step is to add the _ user ID field. Second step is to make a correction rule: name cannot be empty. Third step is to output the log to -- temp directory. Also for priority: 1 High, 2 Medium. That's basically it! Thanks!
Punctuation from voice (colon, --)Command recognition (_)List formattingRemoved fillers
Listening for your voiceMic ready
appsActive appSlack
graphic_eqTranscript Transcribing
“Um... check the API spec for me, first step is to add the. underscore. user ID field, second step is to make a. correction. rule. colon. name cannot be empty. third step is to output the log. to. double dash. temp directory. Priority. circle one. high. circle two. medium. That's basically it! Thanks!”
translateTranslation Assistantauto_awesomeSmart Enhance
Scribis Precision AI Workspace
Your all-in-one studio for transcription, dubbing, and high-precision editing. Built for professionals who demand millisecond accuracy.
record_voice_overOne-click Synthesis Pro
faceSpeaker Identified
translateTranslation
movieHD 1080pLIVE PREVIEW
subtitlesCaptionsDialogueView
We routinely feed them entire novels, dense code bases, and video in a single prompt.
Universal API Connectivity
Scribis is your bridge to the AI ecosystem. Plug in your own API keys for cloud-scale power, or deploy private models locally.
- dynamic_feedCloud & Local HybridNative support for OpenAI, Gemini, and OpenRouter, alongside local GGUF running via Llama.cpp.
- speech_to_textNext-Gen TTSSynthesize results with Kokoro TTS or exclusive experimental local Taiwan voice models.
API Settings
HYBRID RUNTIMEsmart_toyOpenAI API Key
CONNECTED (GPT-5.6, Whisper)auto_awesomeGoogle Gemini API Key
CONNECTED (Gemini 3.1 Pro)hubOpenRouter API Key
READY (200+ Models)Custom Endpoints
add_circle_outline+ Add Custom Endpoint
Real workflows
From raw audio to ready-to-share
Scribis goes beyond transcripts: verify text against a script and on-screen captions, translate subtitles, synthesize speech from a speaker reference, and prepare the finished files for delivery.
LOCAL FIRST
Run transcription on your device
Choose an on-device speech model to process audio locally, with cloud services available when you need them.
BATCH TRANSCRIPTION
Process a whole recording queue
Drop in multiple files or a folder, track each transcription, and skip media that has already been processed.
PERSONALIZED CORRECTION
Your terminology, recognized correctly
Build replacement rules and pronunciation entries using IPA, Bopomofo, or Pinyin to improve recognition of specialist terms.
SPEAKER RECOGNITION
Recognize the voices you know
Organize speaker groups and voice samples to help identify familiar people in future recordings.
CAPTION EDITING
Edit video with your transcript
Align text to the waveform, rough-cut from the transcript, add zooms or mosaics, then export captions or a finished video.
VIDEO RESEARCH
Turn YouTube into a searchable library
Load a video and extract or translate its captions. With Gemini configured, ask questions, create timestamped summaries, and analyze selected scenes.
AI TOOL INTEGRATION
Connect Scribis to your AI toolkit
Use the local MCP server with Claude Code, Codex, Cursor, and other tools to read subtitles, trigger transcription, and work with the timeline.
VOICE ENHANCEMENT
Suppress noise and bring speech forward
Optionally download a 48 kHz neural denoising model to reduce background noise and enhance voice in difficult recordings.
REFERENCE VOICE SYNTHESIS
Carry a speaker’s tone into translated speech
Use a voice sample from the video to synthesize translated lines in a similar voice. A supported speech model and the speaker’s consent are required.
ON-SCREEN TEXT
Check transcripts against video captions
Select the caption area in a video frame, recognize text across frames, and compare it with speech recognition to spot lines that need review.
MULTILINGUAL SUBTITLES
Translate, edit, and export in one flow
Translate and edit subtitles in the same workspace, adjust timings and speakers, then preview bilingual captions or export subtitles and video.
SCRIPT VERIFICATION
Check the transcript against your script
After transcription, compare the source manuscript with subtitle segments, review unmatched lines, edit suggested text, and confirm before opening the translation workspace.