# High-Accuracy Speech Recognition

NextVoice is an open-source speech-to-text solution developed by Lamapi, supporting high-quality transcription and real-time processing.

Medical scenario-specific speech recognition model fine-tuned based on whisper-large-v3

Wav2Vec2 large speech recognition model fine-tuned on the TEDLIUM corpus, supporting English speech-to-text conversion

Wav2Vec2 is a speech recognition model developed by Facebook. It learns speech representations from raw audio through self-supervised learning and is fine-tuned on the LibriSpeech dataset to achieve high-accuracy speech transcription.