insanely-fast-whisper
Control speaker count in insanely-fast-whisper using min-speakers and max-speakers parameters for flexible diarization. Automatically discover optimal speaker numbers within your specified range.
Internal Configuration of the Transformers Pipeline in Insanely‑Fast‑WhisperExplore the internal configuration of the insanely-fast-whisper Transformers pipeline. Optimize ASR with float16, FlashAttention 2, and device-specific settings for faster performance.
How to Provide a HuggingFace Token for pyannote.audio Access in insanely-fast-whisperLearn how to provide your HuggingFace token to insanely-fast-whisper using the --hf-token argument. Securely access pyannote.audio for speaker diarization.
How to Get Word-Level Timestamps in Insanely-Fast-Whisper for Precise TimingUnlock precise timing with word-level timestamps in insanely-fast-whisper. Simply use the --timestamp word CLI flag for accurate audio analysis. Get started now!
Tuning Batch Size for Specific GPU Memory Constraints in Insanely-Fast-WhisperOptimize Whisper GPU memory with precise batch size tuning. Calculate your max parallel processing limit using torch.cuda.max_memory_allocated() and 85% of your VRAM for peak performance.
BetterTransformer Integration in insanely-fast-whisper: A 5× Speedup Deep DiveDiscover how BetterTransformer integration in insanely-fast-whisper achieves 5x faster inference by optimizing compute graphs and fusing attention kernels on CUDA devices without Flash Attention 2.
How to Fix "Torch not compiled with CUDA" Error on Windows with insanely-fast-whisperFix the Torch not compiled with CUDA error for insanely-fast-whisper on Windows. Resolve PyTorch GPU memory issues and get your Whisper model running on the GPU quickly.
How to Select Different Whisper Model Variants (large-v3, distil-large-v2) with insanely-fast-whisperEasily select Whisper model variants like large-v3 or distil-large-v2 with insanely-fast-whisper using the model-name flag or Python API for flexible transcription options.
How to Specify a Custom Output File Path for Transcriptions in insanely-fast-whisperLearn to specify a custom output file path for transcriptions in insanely-fast-whisper using the --transcript-path argument. Control where your transcription results are saved with ease.
How to Install flash-attn in a pipx Environment for insanely-fast-whisperInstall flash-attn in pipx for insanely-fast-whisper hardware acceleration. Boost inference speed with this quick setup guide.
Using the Translate Task Instead of Transcribe in Insanely-Fast-Whisper: A Complete GuideLearn how to use the translate task in insanely-fast-whisper with the --task translate flag. Get translated text instead of transcripts while keeping the same JSON output. Optimize your audio translation workflow.
Resolving CUDA Out-of-Memory Errors with Insanely-Fast-Whisper: Optimization GuideFix CUDA out of memory errors in insanely-fast-whisper by lowering batch size, enabling Flash Attention 2, and optimizing your device. Transcribe audio without crashes.
Have a question about this repo?
These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:
curl -s "https://instagit.com/install.md" 
Maintain an open-source project? Get it listed too →