Transcribe audio with WhisperX, adding word timestamps and speaker labels.
npx skills add https://github.com/ThePlasmak/whisperx --skill whisperx
Please help me install this Agent Skill. Skill: whisperx Source: https://github.com/ThePlasmak/whisperx/tree/main Command: npx skills add https://github.com/ThePlasmak/whisperx --skill whisperx
This Skill automates the process of converting spoken audio into written text, providing highly accurate transcriptions with advanced features like word-level timestamps and speaker identification.
Use the whisperx skill to transcribe the audio file 'meeting_recording.mp3' and identify the speakers.
High-intent search queries and answers about installing and using this skill.
Speech-to-text transcription with word-level timestamps and speaker diarization is handled by the WhisperX model. It provides precise timing for each word via forced alignment and identifies different speakers within the audio.
Generating SRT and VTT subtitle files from spoken audio is fully supported. The transcription process creates these formats for easy integration with video content, alongside JSON output.
Running speech-to-text batched inference requires Python 3.10+ and ffmpeg. An NVIDIA GPU with CUDA is optional but necessary for optimal high-speed processing performance.
Identifying speakers in a meeting recording is best handled through speaker diarization. This feature labels different speakers within the audio while providing accurate transcriptions.
Forced alignment works for creating karaoke-style subtitles by providing word-level timestamps. This precise timing for each word enables exact synchronization between audio playback and text display.
ciTranscribes local audio files to text using NVIDIA Parakeet (NeMo).
maxgent-aiTranscribe audio and video files with word-level timestamps using WhisperX.
steipeteTranscribe audio files into text using the OpenAI Whisper API.
imMamdouhaboammarDesign speech transcription pipelines with local Whisper and cloud ASR services.
