YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context
Rust agentsagiaiai-memoryaudio-recordingcomputer-visionhermeshermes-agentllmlocal-ailocal-firstmachine-learningmcpmultimodalopenclawprivacyrewindscreen-recordingspeech-to-textycombinator
fix(app): skip whisper model pre-download for non-whisper engines
Louis Beaumontd7054041ed63
diff --git a/apps/screenpipe-app-tauri/src-tauri/src/main.rs b/apps/screenpipe-app-tauri/src-tauri/src/main.rsindex 716390829..444ebdf4b 100644--- a/apps/screenpipe-app-tauri/src-tauri/src/main.rs+++ b/apps/screenpipe-app-tauri/src-tauri/src/main.rs@@ -1315,6 +1315,11 @@ async fn main() {// Determine which whisper model the user's config needslet engine = match store_for_download.recording.audio_transcription_engine.as_str() {"deepgram" | "screenpipe-cloud" => None, // Cloud engines don't need local model+ // Non-whisper local engines (parakeet MLX, qwen3) download their own+ // models at load time — don't fetch the 834MB whisper file for them.+ // If the user later switches to a whisper engine, TranscriptionEngine::new+ // downloads it in the background ("will retry at server start").+ "disabled" | "parakeet" | "parakeet-mlx" | "qwen3-asr" => None,_ => {use screenpipe_audio::core::engine::AudioTranscriptionEngine;Some(std::sync::Arc::new(match store_for_download.recording.audio_transcription_engine.as_str() {