Silero Models: pre-trained speech-to-text, text-to-speech and text-enhancement models made embarrassingly simple
#计算机科学#Open Source AI Benchmarking toolkit for benchmarking speech to text services
#自然语言处理#🐍📦 Ultra-fast Python package for calculating and analyzing the Word Error Rate (WER). Built for the scalable evaluation of speech and transcription accuracy.