> ## Documentation Index > Fetch the complete documentation index at: https://docs.zoice.com/llms.txt > Use this file to discover all available pages before exploring further. # AI Video to Text Transcription > Transcribe your videos into accurate text with Zoice's AI transcription tool. Supports 100+ languages, speaker identification, and multiple export formats like SRT and VTT. ## Quick Start Convert spoken content in videos into written text with high accuracy. Navigate to the new transcribe section in your dashboard. Navigate to New Transcribe Choose between YouTube, Instagram, or a media file from your device. Select Link Type Choose the AI model you want to use for transcription. Select Model Paste the video URL in the provided box (if applicable). Paste URL Click the **Generate** button to start the process. Click on Generate ## Supported Video Formats ### File Formats * **MP4**: Most common, recommended * **MOV**: Apple QuickTime * **AVI**: Windows video * **MKV**: Matroska video * **WebM**: Web video format * **FLV**: Flash video ### Specifications * **Max File Size**: 2GB per file * **Max Duration**: 3 hours per video * **Resolution**: Any (audio quality matters most) * **Audio**: Any codec, stereo or mono ## Language Support ### Supported Languages Zoice transcription supports 100+ languages including: * English (US, UK, Australian, Indian) * Spanish (Spain, Latin America) * French, German, Italian, Portuguese * Mandarin, Cantonese, Japanese, Korean * Hindi, Arabic, Russian, Turkish * And many more... ### Auto-Detection Enable automatic language detection for: * Multilingual videos * Unknown language content * Mixed language segments ## Transcription Options ### Timestamp Formats ```srt theme={null} 1 00:00:00,000 --> 00:00:03,500 Welcome to our tutorial on AI transcription. 2 00:00:03,500 --> 00:00:07,000 Today we'll learn how to transcribe videos. ``` **Best For:** Subtitles, video players ```vtt theme={null} WEBVTT 00:00:00.000 --> 00:00:03.500 Welcome to our tutorial on AI transcription. 00:00:03.500 --> 00:00:07.000 Today we'll learn how to transcribe videos. ``` **Best For:** Web videos, HTML5 players ```text theme={null} [00:00:00] Welcome to our tutorial on AI transcription. [00:00:03] Today we'll learn how to transcribe videos. [00:00:07] Let's get started with the basics. ``` **Best For:** Documentation, notes ```text theme={null} Welcome to our tutorial on AI transcription. Today we'll learn how to transcribe videos. Let's get started with the basics. ``` **Best For:** Articles, blog posts, SEO ### Advanced Features Automatically detect and label different speakers: ```text theme={null} [Speaker 1]: Welcome everyone to today's meeting. [Speaker 2]: Thanks for having me. [Speaker 1]: Let's begin with the agenda. ``` * Detects up to 10 speakers * Labels speakers automatically * Option to name speakers manually Automatic punctuation and capitalization: * Periods, commas, question marks * Proper capitalization * Paragraph breaks * Number formatting Options for handling profanity: * **None**: Transcribe as-is * **Asterisks**: Replace with f\*\*\* * **Remove**: Omit profane words * **Tag**: Mark but don't censor Add custom words for better accuracy: * Company names * Product names * Technical terms * Industry jargon * Proper nouns ## Accuracy Optimization ### For Best Results * Clear audio without background noise * Good microphone quality * Minimal echo or reverb * Consistent volume levels * Clear pronunciation * Moderate speaking pace * Minimal overlapping speech * Reduced background conversations * Select correct primary language * Add custom vocabulary * Specify dialect if relevant * Use auto-detect for mixed languages * Specify content type (interview, lecture, etc.) * Enable speaker detection if needed * Set appropriate timestamp density * Choose relevant formatting ## Use Cases ### Subtitle Creation Generate subtitles for videos: ``` Output: SRT or VTT format Timestamps: Every 2-5 seconds Speaker ID: Optional Use: Upload to YouTube, Vimeo, etc. ``` ### Meeting Transcripts Transcribe business meetings: ``` Output: Plain text with timestamps Speaker ID: Enabled Formatting: Paragraphs by speaker Use: Meeting notes, documentation ``` ### Content Repurposing Convert video to blog posts: ``` Output: Plain text only Formatting: Paragraphs Editing: Manual cleanup Use: Blog articles, social posts ``` ### Accessibility Make content accessible: ``` Output: SRT with descriptions Timestamps: Detailed Speaker ID: Enabled Use: Closed captions, accessibility compliance ``` ## Export Formats ### Text Formats * **TXT**: Plain text * **DOCX**: Microsoft Word * **PDF**: Portable document * **MD**: Markdown format ### Subtitle Formats * **SRT**: SubRip subtitles * **VTT**: WebVTT subtitles * **ASS**: Advanced SubStation Alpha * **SSA**: SubStation Alpha ### Data Formats * **JSON**: Structured data with metadata * **CSV**: Spreadsheet format * **XML**: Extensible markup ## Post-Processing ### Built-in Editor Edit transcripts before export: * Fix errors or typos * Adjust timestamps * Merge or split segments * Add speaker names * Format text ### Batch Operations * Find and replace * Bulk timestamp adjustment * Speaker renaming * Format conversion ## Pricing & Credits ### Credit Usage * **Standard Quality**: 1 credit per minute * **High Accuracy**: 2 credits per minute * **Speaker ID**: +0.5 credits per minute * **Rush Processing**: 2x credits ### Processing Time * **Standard**: 1x video length * **Fast**: 0.5x video length * **Rush**: 0.25x video length ## Common Scenarios Create captions and transcripts Generate show notes and transcripts Transcribe interviews with speaker ID Create study materials from lectures ## Troubleshooting * Check audio quality * Add custom vocabulary * Verify language selection * Use high accuracy mode * Reduce background noise * Increase audio volume * Add words to custom vocabulary * Check for audio dropouts * Try high accuracy mode * Manually correct in editor * Ensure distinct voices * Check microphone setup * Reduce voice overlap * Adjust timestamp density * Check video sync * Manually adjust in editor * Verify frame rate For best results, use videos with clear audio and minimal background noise. Consider using a noise reduction tool before transcription if needed. ## Privacy & Security * **Encrypted Upload**: All videos encrypted in transit * **Secure Storage**: Temporary storage, auto-deleted after 24 hours * **No Training**: Your content is never used for model training * **Confidential**: Transcripts are private to your account ## Next Steps Convert text back to professional speech Create video presentations with AI characters Explore the main Voice Generator features