wav2vec2-live and wav2asr
One tool provides a live speech recognition application utilizing Wav2vec 2.0, while the other offers a library-level implementation of the Wav2vec 2.0 framework for ASR tasks, making them ecosystem siblings where the library could serve as a foundational component for building applications like the live one.
About wav2vec2-live
oliverguhr/wav2vec2-live
A live speech recognition using Facebooks wav2vec 2.0 model.
Provides real-time streaming speech recognition by continuously processing microphone input through any wav2vec2 model from Hugging Face, with configurable audio devices and per-inference timing metrics. The architecture uses PyAudio for live audio capture and runs inference asynchronously, returning recognized text alongside processing latency and sample duration for performance monitoring.
About wav2asr
phanxuanphucnd/wav2asr
A library version of wav2vec 2.0 framework for Automatic Speech Recognition task.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work