- expanded stretch ratios from 0.25X to 4.00X by cascading instances
- new feature to apply different ratio to detected silence/gaps
- addition of sample audio files and build/test scripts
- reduced memory usage and improved performance
- fix for numeric overflow in very loud audio
- improved portability (tested on ia16-gcc)
Many thanks to Hayati Ayguen for suggestions and patches!