User Guide
Follow these steps to easily complete voice cloning and speech generation
1. Upload Audio Sample
Select a clear audio file (MP3 or WAV format), ensuring good sound quality without background noise. Recommended duration is 10-30 seconds.
Tips:
- Choose audio with clear pronunciation and natural emotion
- Avoid background music or noise
- File size should not exceed 50MB
2. Configure Training Parameters
Select the appropriate training model type, enter the model name and reference text. The reference text should match the audio content.
Tips:
- S1 model is suitable for most scenarios
- Model name should be distinctive
- Reference text must match audio content exactly
3. Start Training
Click the "Start Training" button, and the system will automatically analyze audio features and train the AI model. Training usually takes 1-3 minutes.
Tips:
- Keep the page open to avoid interruption
- Model will be saved automatically after training
- View all models in "My Models"
4. Generate Speech
After training, go to the "Text to Speech" page, select the trained model, and input the text you want to convert to speech.
Tips:
- Use punctuation properly for more natural speech
- Recommend no more than 1000 characters per generation
- Generated audio can be downloaded and saved