Research: The LYRICOS Singing Voice Synthesis System

GOAL: Synthesize natural-sounding singing voice from MIDI file input

In a collaborative effort between Mike Macon and Mark Clements of the Center for Signal and Image Processing, Leslie Jensen-Link and James Oliverio of the Georgia Tech Music Department, and E. Bryan George of Texas Instruments, a system capable of synthesis of high-quality singing voice from a MIDI file input was developed.

Demonstrates synthesis capabilities:

• selection of 1-3 phone units
• sinusoidal model synthesis
• MIDI graphical interface
• musical controls (vibrato, etc.)

Demonstrations of the LYRICOS system

These examples were synthesized using a MIDI file input. A sinusoidal model analysis was performed on a single recording of the syllables "/L/AA/","/L/ /EY/",. From these prototype syllables, the sinusoidal model performs the synthesis of the musical phrase based on control functions specified by the MIDI file.

• Female human au, wav
• Resynthesis with pitch, duration modification to create a new melody au, wav
• Three part harmony au, wav

We have also extended the synthesis capability to cover any arbitrary English utterance. The next example was synthesized using a set of phonetic transition units stored in an online inventory. The system selects several of these units at runtime, concatenates them, and then performs smoothing and pitch/time-scale modification. For more details, see our paper presented at ICASSP 1997 in Munich.

• Male human singing nonsense words from database au, wav
• Synthesis from MIDI: Frere Jacques au, wav

Also you might want to check out our new system for singing voice synthesis.