VibeVoice-ASR: speech-to-text model designed to handle 60-minute long-form audio(huggingface.co)
9 points by maxloh 21 hours ago | discuss