1 Commits
Author SHA1 Message Date
kcfancherandGitHub 63496d05a2 Fix transcription sample-rate handling (#740)
Fix for #703.

Adds an optional sampleRate parameter to the decodeAudioToFloat32() function. The caption transcriber passes in a value of 16000 to prepare audio in the format Whisper expects.
2026-03-28 19:06:40 +01:00