pengzhendong/streaming-sensevoice

Issues

• 识别结果中许多词语连起来了 如何解决

  #27 opened 8 days ago
  1

• 实时推理中的标点恢复

  #26 opened 16 days ago
  2

• 8000HZ PCM音频流VAD返回开启和关闭

  #25 opened 18 days ago
  3

• Offline mode inference

  #24 opened 18 days ago
  1

• 调用websocket接口时性能消耗问题

  #23 opened 24 days ago
  1

• 依照 Issues #21 修改L130后报错

  #22 opened a month ago
  1

• [Question] 如何设定其他语言？

  #21 opened a month ago
  2

• [FEATURE] 如何实现实时更新热词

  #20 opened a month ago
  1

• [BUG] `timestamp` 长度与字符串长度不匹配

  #19 opened a month ago
  0

• [FEATURE] Websocket demo 实地测试中暴露出的问题

  #18 opened a month ago
  2

• 执行python main.py时产生的一个报错，但无法定位位置

  #16 opened a month ago
  1

• [Feature] 为 `StreamingSenseVoice` 添加静态方法 `load_model` 以便预加载模型

  #14 opened a month ago
  1

• [Question] 请问热词方案是否能处理硬造的英文词？

  #11 opened a month ago
  1

• [BUG] [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int32)) , expected: (tensor(int64))

  #10 opened a month ago
  6

• [Information] RTF 测量结果

  #8 opened a month ago
  0

• [Feature] 是否有计划支持 Batch 推理？

  #7 opened a month ago
  1

• 请问这种实现方式对情绪和事件识别有什么影响？

  #6 opened a month ago
  1

• 如何采用麦克风语音实时输入，流式ASR输出呢？

  #5 opened a month ago
  1

• 如何使ASR结果带有标点符号？

  #4 opened a month ago
  1

• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument index in method wrapper_CUDA__index_select)

  #3 opened a month ago
  1

• INVALID_ARGUMENT : Unexpected input data type. Actual: (tensor(int32)) , expected: (tensor(int64))

  #2 opened a month ago
  1