The user is discussing an AI voice input project and its performance, mentioning a score of 0.88. They suggest improvements by ensuring GPU usage through automatic configuration and cloud storage management for training files. The user is considering saving the best model based on loss values or by rounds, and is contemplating how to share these files among multiple accounts for collective training. They also mention the use of web disk services like WebDF for storage and access. The user expresses excitement about the project's potential and its integration with technologies like Groq Whisper for transcription.

标签: none

评论已关闭