modelscope/FunASR

Issues

• funasr onnx推理sensevoice的时候能否解除torch依赖

  #2253 opened 25 days ago by ILG2021
  1

• paraformer模型在进行中英文混合训练后，英文全部变为大写且没有空格

  #2244 opened a month ago by Promethues3
  1

• 如何在实际应用中提升模型效率？

  #2252 opened a month ago by mzgcz
  0

• 请问，像那种专业词汇量大的场景下，推荐使用funasr么？使用语言模型，目前微调不了么？

  #2251 opened a month ago by qqqq640
  4

• 使用funasr的时候，可以挂一个语言模型么

  #2250 opened a month ago by qqqq640
  1

• vad_model和punc_model设置本地路径报错

  #2248 opened a month ago by LIUKAI0815
  1

• 请问Paraformer-V2的代码会开源吗？

  #2208 opened a month ago by NiniAndy
  15

• 'language-score-prediction is not in the pipelines registry group language-score-prediction. Please make sure the correct version of ModelScope library is used.'

  #2247 opened a month ago by seetimee
  0

• 在阿里云服务器容器内部署了ASR系统，本机打开html5输入wss://服务器ip:10096 ，报错连接服务器失败

  #2243 opened a month ago by AliceShen122
  1

• FunASR开源项目体验demo中路径runtime/python/websocket/funasr_wss_server.py中没有timestamp时间戳，在线体验中的时间戳从何而来？

  #2195 opened 2 months ago by mrblacklee
  2

• Funasr VAD GetFrameState index is out of bounds.

  #2242 opened a month ago by momaek
  0

• 如何提高微调seaco_paraformer模型时的GPU利用率

  #2232 opened a month ago by JVfisher
  1

• paraformer微调之后模型变大，且和basemodel推理同一段wav文件时会报错

  #2239 opened a month ago by YouTwoMeToo
  1

• math.exp can lead to overflow

  #2241 opened a month ago by tshmak
  0

• mispelt "epoch" led to bug in save/loading checkpoints

  #2240 opened a month ago by tshmak
  0

• 编译websocket的gpu版本是否一定需要torch_blade?

  #2238 opened a month ago by Nuomanzzz
  0

• funasr-runtime-sdk-online-cpu-0.1.12镜像的crash问题

  #2237 opened a month ago by MyWestCity
  0

• Docker容器版本funasr-runtime-sdk-gpu-0.2.0返回结果没有stamp_sents字段只有timestamp字段

  #2233 opened a month ago by caoshiny
  1

• FunASR是否支持实时或离线解析话务8K采样率

  #2236 opened a month ago by jgjjgjjhhg
  0

• speaker darization clustering result is very bad for short audio

  #2235 opened a month ago by MonolithFoundation
  0

• 使用speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online-onnx这个模型推理不出来东西

  #2234 opened a month ago by chwyxy
  0

• Problems to use vad_model, pucn_model and spk_model with streaming voice. 如何正常在流式处理中加载这3个模型?

  #2231 opened a month ago by 1113200320
  0

• Adds option to disable installation from requirements.txt

  #2230 opened a month ago by tshmak
  2

• Array index out of bounds causes segment fault when ASR result is empty

  #2227 opened a month ago by keeofkoo
  1

• 是否有模型再分割两个说话人耦合的语句？

  #2229 opened a month ago by lucasjinreal
  0

• VAD 流式推理的问题

  #2228 opened a month ago by TungyuYoung
  0

• where to get the onnx model for paraformer triton

  #2224 opened a month ago by didadida-r
  0

• sensevoice convert onnx to triton fail

  #2226 opened a month ago by didadida-r
  0

• Arbitrary code execution on this line of code

  #2225 opened a month ago by tshmak
  0

• UnboundLocalError: local variable 'get_tokenizer' referenced before assignment

  #2223 opened a month ago by johnhula
  0

• PCM录音文件读取似乎有问题

  #2207 opened a month ago by HandsomeJM
  2

• vad 中最后一个字的时间戳判断不准确。

  #2222 opened a month ago by wujpbb7
  0

• cannot import name 'AutoModel' from partially initialized module 'funasr' (most likely due to a circular import)

  #2221 opened a month ago by H4M5TER
  1

• pyinstaller6.11版本打包后运行报错

  #2220 opened a month ago by SFidea
  0

• cpu离线版本，客户端建立websocket请求，串行推送多个48k的语音wav，语音转写服务会重启，而且概率很高，log.txt也没见具体报错原因。生成的dmp（每次5g左右）文件会导致docker容器不断的变大。

  #2204 opened a month ago by janchou92
  3

• VAD效果很差，是使用问题？

  #2217 opened a month ago by young1013
  2

• python websocket客户端链接服务端不返回识别结果

  #2215 opened a month ago by yichuxue
  1

• 部署在docker容器内的SensevoiceSmall模型该如何接收入参？

  #2214 opened a month ago by lin-xiaosheng
  0

• 为什么读取本地音频报错

  #2213 opened a month ago by deict
  0

• 为什么读取本地音频报错

  #2212 opened a month ago by deict
  0

• IndexError: index 3 is out of bounds for dimension 1 with size 2

  #2211 opened a month ago by JonneryR
  2

• 使用iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch模型，运行offline_gpu服务报错

  #2200 opened 2 months ago by gzqqqqqq
  2

• Intel MKL function load error: mkl_vml_kernel_sLn_ttab.

  #2210 opened a month ago by 1749352011
  0

• funasr有没有类似于faster_whisper的whisperModel().transcribe的功能，可以获取segments?

  #2198 opened 2 months ago by liuchangzong
  1

• FunASR 语音识别中的回声问题？

  #2203 opened a month ago by JACKYLUO1991
  0

• vad FSMN语音端点检测-中文-通用-16k 内存泄漏问题

  #2202 opened a month ago by tonyzzzzz
  0

• GPU版找不到文件model_blade.torchscript

  #2201 opened 2 months ago by Jotree2012
  2

• 使用damo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404模型，运行offline_gpu服务报错

  #2199 opened 2 months ago by gzqqqqqq
  0

• vad 模式切割的音频在转录后会自己删除吗

  #2197 opened 2 months ago by lesrose
  0

• docker服务端发送文本内容过长，超出缓冲区，导致消息发送失败

  #2196 opened 2 months ago by psk-github
  0