Sync Model

Youka offers different AI models for synchronizing lyrics with music. Each model has its strengths.

Model Types #

• Alignment models require you to provide the lyrics. They match your lyrics to the audio timing.
• Transcription models automatically detect and transcribe lyrics from the audio, then sync them.

Available Sync Models #

Audioshake Alignment (Default) #

• Works with most music genres
• Good balance of speed and accuracy
• Requires lyrics to be provided

Best for: Pop, rock, and mainstream music when you have lyrics

MusicAI Alignment #

• Alternative alignment model
• Works well for various genres
• Requires lyrics to be provided

Best for: When Audioshake doesn't give optimal results

Audioshake Transcription #

• Automatically transcribes lyrics from audio
• No need to provide lyrics manually
• High accuracy for English songs

Best for: When you don't have the lyrics

MusicAI Transcription #

• Alternative transcription model
• Supports multiple languages
• Automatic lyrics detection

Best for: Non-English songs or alternative transcription

Whisper AI #

• Advanced speech recognition by OpenAI
• Supports multiple languages
• Excellent for non-English songs

Best for: International music, multilingual content

Choosing a Sync Model #

When editing lyrics:

1. Go to Edit > Edit Lyrics
2. Enter or edit your lyrics (for alignment models)
3. Select your preferred sync model from the dropdown
4. Click Save to start the sync process

Tips for Best Results #

Music Type	Recommended Model
English Pop (have lyrics)	Audioshake Alignment
Unknown Lyrics	Audioshake Transcription
Non-English	Whisper AI or MusicAI Transcription
Complex Timing	Try different models, then use Manual Sync to fine-tune