AIGCLINK2025-02-20 16:14:11西北工业大学开源了一个语音理解模型:OSUM,支持8种语音理解任务,从ASR到情感识别,以及语音到文本的深度理解 支持语音识别(ASR) 带时间戳的语音识别(SRWT) 声音事件检测(VED) 语音情感识别(SER) 说话风格识别(SSR) 说话者性别分类(SGC) 说话者年龄预测(SAP) 语音到文本对话(STTC) 它结合了Whisper和Qwen2,采用ASR+X训练策略,支持多任务同时
Nury Vittachi2025-01-22 00:27:35A Chinese computer nerd is stunning tech experts with a bargain price AI tool that beats the world's top players on several benchmarks. Liang Wenfeng's system, called DeepSeek-V3, has powers of "reas