Voice-to-text dictation app with local (Nvidia Parakeet/Whisper) and cloud models (BYOK). Privacy-first and available cross-platform.
JavaScript aianthropiccross-platformgeminigroqlinuxmacosnvidiaopen-sourceopenaiparakeetspeech-to-texttranscribewhisperwindows
fix: reduce audio chunk segment duration to fit Vercel's payload limit
Gabriel Steind70a4bf4f6a1
diff --git a/src/helpers/ipcHandlers.js b/src/helpers/ipcHandlers.jsindex d09463f26..d6f4eac20 100644--- a/src/helpers/ipcHandlers.js+++ b/src/helpers/ipcHandlers.js@@ -2154,7 +2154,9 @@ class IPCHandlers {chunksCompleted: 0,});- const chunkPaths = await splitAudioFile(filePath, chunkDir);+ const chunkPaths = await splitAudioFile(filePath, chunkDir, {+ segmentDuration: 240, // ~3.75 MB/chunk, under Vercel's 4.5 MB payload limit+ });const totalChunks = chunkPaths.length;debugLogger.debug("Audio split into chunks", { totalChunks });