AI Capabilities

TongFlow integrates over 20 AI capabilities spanning generation, editing, and analysis. This guide explains what each feature does and when to use it.

Generation Features

Text Generation

Create written content using large language models.

Powered by: Gemini, DeepSeek, Qwen

Best for:

  • Writing scripts and stories
  • Generating product descriptions
  • Translating content
  • Answering questions and research

Image Generation

Create images from text descriptions.

Powered by: Qwen Image, Nunchaku

Best for:

  • Concept art and illustrations
  • Marketing visuals
  • Product mockups
  • Social media content

Tips: Be specific with style, lighting, and composition for better results.

Video Generation

Create videos from images or text.

Types available:

  • Image to Video: Animate a still image
  • Text to Video: Generate from description
  • First-Last Frame: Create video between two keyframes
  • Speech-driven: Sync video to voice

Best for:

  • Short-form social content
  • Product demos
  • Animated storytelling

Audio Generation

Text to Speech: Convert text to natural-sounding voice

  • Multiple languages and accents
  • Adjustable speed and tone

Text to Music: Generate music from descriptions

  • Various genres and moods
  • Background music and jingles

Voice Cloning: Replicate a voice from samples

  • Preserve unique vocal characteristics
  • Create consistent character voices

Editing Features

Image Editing

Modify existing images with AI assistance.

Capabilities:

  • Instruction-based: Describe changes in natural language
  • Multi-angle: Create consistent views of the same subject
  • Refinement: Enhance details and quality

Image Enhancement

Upscaling: Increase resolution up to 4x

  • Works with photos and illustrations
  • Preserves and enhances details

Segmentation: Intelligent background removal

  • Clean cutouts for product photos
  • Prepare assets for compositing

Video Editing

Subtitle Removal: Clean text overlays from videos

  • Preserves background content
  • Works with burned-in subtitles

Watermark Removal: Remove unwanted logos

  • Smart content reconstruction
  • Maintains video quality

Upscaling: Enhance video resolution

  • Improve older or low-quality footage

Analysis Features

Image Understanding

Extract information from images.

Capabilities:

  • Describe image contents
  • Identify objects and scenes
  • Read text from images (OCR)
  • Answer questions about images

Video Understanding

Analyze video content.

Capabilities:

  • Summarize video content
  • Identify scenes and actions
  • Generate descriptions

Speech Recognition

Convert spoken audio to text.

Capabilities:

  • High-accuracy transcription
  • Multiple language support
  • Timestamps for subtitling
  • Speaker identification

Document Analysis

Extract content from documents.

Supported formats: PDF, images with text

Capabilities:

  • Text extraction
  • Layout preservation
  • Table recognition

Audio Processing

Noise Reduction

Clean up audio recordings.

  • Remove background noise
  • Improve voice clarity

Track Separation

Split audio into components.

  • Separate vocals from music
  • Extract individual instruments

Voice Conversion

Transform voice characteristics.

  • Change pitch and tone
  • Apply different voice styles

Social Media Integration

Import content from social platforms.

Supported platforms:

  • TikTok
  • Douyin (抖音)
  • Instagram
  • Xiaohongshu (小红书)
  • Kuaishou (快手)

What’s extracted:

  • Video files
  • Audio tracks
  • Captions and descriptions

Usage Tips

  1. Combine capabilities: Chain multiple AI features for complex workflows
  2. Iterate: Run multiple times with refined prompts for better results
  3. Check outputs: AI can make mistakes — review before publishing
  4. Be specific: Detailed prompts produce more accurate results

Next Steps

  • Materials — Manage your generated content
  • Account — Understand credits and subscriptions