mirror of
https://github.com/heygen-com/hyperframes.git
synced 2026-09-05 00:56:23 +00:00
* fix(render): normalize local AAC duration before mux * style(render): apply repository formatter * fix(render): count AAC packets for duration normalization Older FFmpeg versions estimate raw ADTS duration from bitrate and can undercount variable-bitrate audio, causing the normalizer to append a false silence tail. Derive the mixed AAC duration from packet count and sample rate instead. * fix(render): isolate normalized audio temp path * fix(engine): centralize AAC packet duration * test(producer): refresh AAC duration golden