How to Use
• Upload audio files and type their transcript in the Source section, or choose a preset like Michael Rosen • Type the text you want to create in the Destination transcript box • Click Submit to generate audio that makes the source voice "speak" your destination text • The tool matches phonemes (speech sounds) from the source audio to reconstruct new words • Wait for processing - first generation can take several minutes • Adjust options to fine-tune results: - Match settings control how strictly phonemes are matched - Influence sliders affect which audio segments are selected - Correction options transfer pitch, volume, and timing • Download results as Adobe Audition sessions or JSON data • Use the microphone button to record audio directly in your browser • Upload existing transcripts as JSON or TextGrid files to skip manual typing