speaker embedding
A mathematical vector representation of a speaker's voice characteristics, created by a neural network model, that can be used to identify and distinguish speakers
As defined in OpenAI Whisper Speaker Diarization - Transcription with Speaker Names
Where it comes up · 1
Most important claims · 3 of 3Search all claims for “speaker embedding” →
- After Whisper transcribes the audio into segments with timestamps, the pipeline creates speaker embeddings using the loaded SpeechBrain embedding model for each audio segment.OpenAI Whisper Speaker Diarization - Transcription with Speaker Names
- After creating speaker embeddings, the pipeline uses agglomerative clustering (from scikit-learn) to group embeddings by speaker, with the number of clusters determined by the user-specified speaker count.OpenAI Whisper Speaker Diarization - Transcription with Speaker Names
- The speaker diarization pipeline requires installation of two libraries: Whisper (for transcription) and Pyannote (for annotation), and imports the SpeechBrain speaker embedding model (ECAPA TDNN embeddings trained on Voxceleb) for speaker verification.OpenAI Whisper Speaker Diarization - Transcription with Speaker Names