Speaker Identification : Recognizing who is speaking based on their voice characteristics.
  • Python 99.4%
  • Shell 0.6%
Find a file
2025-06-01 23:23:54 +02:00
hparams feat: batch size + workers + get val dataset 2025-05-28 00:20:18 +02:00
utils feat: add m4a to wav util file 2025-05-18 18:17:34 +02:00
.gitignore feat: training with Huggign face CV dataset 2025-05-26 00:19:22 +02:00
common_voice_prepare.py feat: add datasets prepare files from speechbrain 2025-05-18 18:16:46 +02:00
common_voice_utils.py feat: batch size + workers + get val dataset 2025-05-28 00:20:18 +02:00
extract_speaker_embeddings.py feat: add speacker verif from speechbrain 2025-06-01 23:15:22 +02:00
README.md docs: add Acknowledge to speechbrain 2025-05-18 18:16:14 +02:00
requirements.txt feat: add requirements 2025-05-26 00:20:41 +02:00
setup_runpod.sh feat: add oneliner 2025-06-01 23:23:54 +02:00
speaker_verification_cosine.py feat: add speacker verif from speechbrain 2025-06-01 23:15:22 +02:00
train_cv.py feat: switch to fr dataset 2025-05-26 00:43:11 +02:00
train_speaker_embeddings.py feat: add missing file 2025-05-26 00:32:58 +02:00
voxceleb_prepare.py feat: add datasets prepare files from speechbrain 2025-05-18 18:16:46 +02:00

🎙️ VoiceID

VoiceID is a deep learning-based speaker identification system that recognizes people based on their unique voice characteristics. The system can identify speakers from a known set of registered voices using audio recordings.

📌 Features

  • Speaker Identification (1:N):

    • Identify the person from a known set of enrolled speakers.
  • Speaker Verification (1:1):

    • Check if a claimed identity matches the voice sample (for authentication use cases).

Acknowledge

This repo is based on the excellent SpeachBrain repository