WhisperKit
WhisperKit copied to clipboard
argmaxinc
On-device Speech Recognition for Apple Silicon
Hello, I’m experiencing an issue with transcription in non-English streams, particularly in Russian, where the number of hypothesis words generated is disproportionately large compared to confirmed words. The hypothesis words...
#124 adds MLX support. We need to adopt the [in-place KV cache update](https://github.com/ml-explore/mlx-swift-examples/blob/main/Libraries/LLM/KVCache.swift#L9) pattern to improve performance.
I’ve been using WhisperKit for real-time stream transcription in a project, and I’ve noticed that as time progresses, particularly after 20-30 minutes of continuous use, the transcription speed begins to...
**Environment:** - MacBook Pro 14 (2021) 16GB - Apple M1 Pro - macOS 14.2.1 (23C71) **Issue Description:** I compiled whisperkit-cli using the whisperkittools pipeline. Subsequently, I noticed that each time...
https://github.com/argmaxinc/WhisperKit/blob/3ebfa142a0e181668882e8e1c54088a528e2907b/Sources/WhisperKit/Core/Audio/AudioProcessor.swift#L416-L426
In OpenAI’s Whisper model, setting the language to “auto” allows for automatic language detection, enabling multilingual transcription. However, in WhisperKit, this “auto” setting doesn’t seem to function as expected. Could...
Hi all, This mp3 file seems to trip WhisperKit (and latest Testflight WhisperAX on macOS) up, just wanted to let you know :) I haven't dived much further though mp3...
When referencing WhisperKit from a SPM package as a dependency, it does not recognize the "v" prefixed version as a version string (for both from and exact). The error in...
The iPhone 12 mini crashed while using the latest version of the WhiserAX app on TestFlight to transcribe an 11-hour audio import. Selected model: whisper-small.en
