Voice_Extractor — инструмент на базе ИИ для извлечения чистых фрагментов речи конкретного говорящего из многоголосых аудиозаписей.
PyTorch, PyAnnote, Whisper, WeSpeaker, Bandit-v2, SpeechBrain, Silero-VAD, FFmpeg и др.
pip install -r requirements.txt
python run_extractor.py \
--input-audio path/to/audio.wav \
--reference-audio path/to/target.wav \
--target-name "SpeakerName" \
--token "hf_yourToken"
#Python #VoiceExtractor #SpeechAI #SpeakerDiarization #Whisper #PyAnnote #WeSpeaker #OpenAI #SpeechToText #AudioProcessing #DeepLearning #SpeechSeparation #MLTools #HuggingFace #AudioAI #AI4Audio #ASR
https://github.com/ReisCook/Voice_Extractor
Please open Telegram to view this post
VIEW IN TELEGRAM
GitHub
GitHub - ReisCook/Voice_Extractor: Automated speech dataset creator
Automated speech dataset creator. Contribute to ReisCook/Voice_Extractor development by creating an account on GitHub.
2 3👍2
voicetag — Python‑библиотека для speaker diarization + named identification: кто говорил и когда в аудио (подкасты, митинги, звонки) — pyannote + resemblyzer в 3 строки кода.
vt.enroll("Alice", ["a1.flac"]); result = vt.identify("meeting.flac") — timeline с confidence + overlap detection.vt.transcribe("audio.flac", provider="whisper") — кто что сказал (OpenAI/Groq/Deepgram/Whisper).voicetag enroll Alice *.flac; voicetag identify meeting.flac + profiles JSON.pip install voicetag[whisper] # + HF token для pyannote
GPU:
device="cuda" или MPS. Language agnostic (EN/HE/ZH). #python #speech #diarization #voicetag #pyannote #transcription
https://github.com/Gr122lyBr/voicetag
Please open Telegram to view this post
VIEW IN TELEGRAM
GitHub
GitHub - Gr122lyBr/voicetag: Speaker identification powered by pyannote and resemblyzer
Speaker identification powered by pyannote and resemblyzer - Gr122lyBr/voicetag
1🔥4👍3⚡1