Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / modelscope/FunASR issues and pull requests
#1776 - 导入funasr包时间过长
Issue -
State: closed - Opened by Erlonealpha 5 months ago
Labels: question
#1774 - 尝试使用 finetune.sh 微调 seaco-paraformer ,微调后发现,每个epoch 的model.pt增大了很大,从原模型的800多M,增大到了2.44GB多,这是什么原因
Issue -
State: closed - Opened by locasxe 6 months ago
- 3 comments
Labels: bug
#1774 - 尝试使用 finetune.sh 微调 seaco-paraformer ,微调后发现,每个epoch 的model.pt增大了很大,从原模型的800多M,增大到了2.44GB多,这是什么原因
Issue -
State: closed - Opened by locasxe 6 months ago
- 1 comment
Labels: bug
#1773 - 调用FunASR/runtime/http服务出现错误 basic_string::_M_construct null not valid
Issue -
State: open - Opened by hooploop 6 months ago
- 6 comments
Labels: question
#1771 - 如何训练标点预测模型
Issue -
State: closed - Opened by LIUKAI0815 6 months ago
#1770 - funasr-runtime-sdk-online-cpu-0.1.10软件包内存一直缓慢涨,不下降
Issue -
State: closed - Opened by MyWestCity 6 months ago
- 5 comments
Labels: bug
#1766 - How to create audio profiles for speakers for re-identification?
Issue -
State: closed - Opened by hmehdi515 6 months ago
Labels: question
#1762 - docker运行使用ssl失败,访问wss://127.0.0.1:10095/不行,ws可以
Issue -
State: closed - Opened by zhouganglong 6 months ago
- 3 comments
Labels: question
#1761 - ValueError: invalid character: @
Issue -
State: closed - Opened by WjMessi1 6 months ago
- 1 comment
Labels: bug
#1756 - 报错 websocketpp.processor:4 (A message was too large)
Issue -
State: open - Opened by dfengpo 6 months ago
- 8 comments
Labels: bug
#1756 - 报错 websocketpp.processor:4 (A message was too large)
Issue -
State: open - Opened by dfengpo 6 months ago
- 9 comments
Labels: bug
#1755 - 基于funasr paraformer-onnx 搭建了一个语音转写网页,感兴趣的同学可以试一试
Issue -
State: open - Opened by pika-online 6 months ago
- 15 comments
Labels: documentation
#1728 - Qwen-Audio + VAD 搭配使用报错
Issue -
State: open - Opened by zhangyucha0 6 months ago
- 2 comments
Labels: bug
#1723 - 使用了GPU运行,但是CPU使用率依旧很高?根据性能报告,GPU确实生效。
Issue -
State: closed - Opened by 12915494174 6 months ago
- 4 comments
#1713 - sentence_info 内容不全
Issue -
State: closed - Opened by tiaanaqiqikuaipao 6 months ago
- 1 comment
Labels: bug
#1710 - asr 模型训练到一半(已训练20轮),突然报错:TypeError: object of type 'int' has no len()
Issue -
State: closed - Opened by Banana-Basilisk 6 months ago
- 5 comments
Labels: question
#1709 - the memory go to high, out of memory
Issue -
State: closed - Opened by huangdadahai2015 6 months ago
Labels: bug
#1708 - memory leak in funasr-runtime-sdk-cpu-0.4.4
Issue -
State: closed - Opened by FD-Liekkas 6 months ago
- 1 comment
Labels: bug
#1701 - Vad模型在进程下数据卡住不动
Issue -
State: closed - Opened by CNXiDaDa 6 months ago
- 1 comment
Labels: bug
#1699 - offline results 少了开头的一个字
Issue -
State: closed - Opened by Skypow2012 6 months ago
- 2 comments
Labels: bug
#1698 - 后处理会抛IndexError异常
Issue -
State: closed - Opened by yzmyyff 6 months ago
Labels: bug
#1683 - 语音中4个字,经常识别丢失了最后一个字,麻烦看下
Issue -
State: closed - Opened by yk3372 7 months ago
- 3 comments
Labels: question
#1674 - 识别结果存在循环输出或输出混乱问题
Issue -
State: closed - Opened by WjMessi1 7 months ago
- 3 comments
Labels: bug
#1673 - 微调模型出现了个问题,请指教我下,哪里弄错了
Issue -
State: closed - Opened by chenmiaotian 7 months ago
- 1 comment
Labels: question
#1673 - 微调模型出现了个问题,请指教我下,哪里弄错了
Issue -
State: closed - Opened by chenmiaotian 7 months ago
- 1 comment
Labels: question
#1668 - 微调会自动删除ep文件, 导致微调结束后找不到需要ep文件
Issue -
State: closed - Opened by bird-9 7 months ago
- 3 comments
Labels: bug
#1660 - auto_model 中punc模型入参为空触发的bug
Issue -
State: closed - Opened by clb-123 7 months ago
- 4 comments
Labels: bug
#1658 - 数组越界问题
Issue -
State: closed - Opened by clb-123 7 months ago
- 7 comments
Labels: bug
#1655 - runtime error for blank audio file
Issue -
State: closed - Opened by jianganghan 7 months ago
Labels: bug
#1648 - 英文batch推理的错误比单个推理多很多
Issue -
State: closed - Opened by wincing2 7 months ago
- 1 comment
Labels: question
#1643 - Vad模型处理数据会一直卡住
Issue -
State: closed - Opened by CNXiDaDa 7 months ago
- 4 comments
Labels: question
#1640 - 请问如何在pipeline中使用外挂的transformer-lm做shallow fusion
Issue -
State: closed - Opened by ZihanLiao 7 months ago
- 1 comment
Labels: question
#1637 - 模型对带有噪声的静音识别会出现随机的文本结果
Issue -
State: closed - Opened by tuocheng0824 7 months ago
Labels: bug
#1632 - README demo: ValueError: not enough values to unpack (expected 3, got 1)
Issue -
State: closed - Opened by flystar8 7 months ago
- 4 comments
Labels: question
#1632 - README demo: ValueError: not enough values to unpack (expected 3, got 1)
Issue -
State: open - Opened by flystar8 7 months ago
- 3 comments
Labels: question
#1629 - 如何对Whisper进行量化
Issue -
State: closed - Opened by ChenjieXu 7 months ago
Labels: question
#1623 - 使用角色识别模型,得到的结果和实际说话人相差太大,有什么提升准确率的参数吗?
Issue -
State: closed - Opened by chenfuckthesky 7 months ago
- 3 comments
Labels: question
#1620 - valgraind 检测 onnxruntime 发现 内存泄漏
Issue -
State: closed - Opened by locasxe 7 months ago
- 2 comments
Labels: bug
#1614 - 中文都支持哪些方言
Issue -
State: closed - Opened by jianking123 7 months ago
- 4 comments
Labels: question
#1613 - 执行后结果为空,不清楚哪里出了问题。
Issue -
State: closed - Opened by jacksinofn 7 months ago
- 1 comment
Labels: question
#1609 - 使用speaker diarization MISS错误率很高,请问是vad模块效果不好吗?还有结合视频的DER结果效果比单音频的还要差,请问这可以微调嘛?
Issue -
State: closed - Opened by Coconut059 7 months ago
- 1 comment
Labels: question
#1606 - 生成标点包含、等时会出现字符长度与时间长度不相等
Issue -
State: closed - Opened by GioGioBond 7 months ago
- 2 comments
Labels: question
#1603 - 用最新的版本,会出现NameError: name 'ClusterBackend' is not defined 报错
Issue -
State: closed - Opened by eleven-monkey 7 months ago
- 7 comments
Labels: bug
#1602 - 直接语音输入比较短的句子会出问题
Issue -
State: closed - Opened by ArboterJams 7 months ago
- 1 comment
Labels: question
#1600 - 求教tokens.json的生成和问题?
Issue -
State: closed - Opened by xiulianzw 7 months ago
Labels: question
#1599 - 最新版funasr,按照教程fineturn 模型报错:forward() missing 4 required positional arguments: 'speech', 'speech_lengths', 'text', and 'text_lengths'
Issue -
State: closed - Opened by Xsx93 7 months ago
- 2 comments
Labels: bug
#1598 - funasr使用qwen-audio支持双卡进行测试?
Issue -
State: closed - Opened by lai-serena 7 months ago
Labels: question
#1597 - 关于使用越南语进行微调,不收敛的问题
Issue -
State: closed - Opened by xiulianzw 7 months ago
Labels: question
#1595 - KeyError: 'asr-inference is not in the pipelines registry group auto-speech-recognition. Please make sure the correct version of ModelScope library is used.'
Issue -
State: closed - Opened by suyufan 7 months ago
- 5 comments
Labels: bug
#1589 - TypeError: unsupported operand type(s) for /: 'list' and 'int'
Issue -
State: closed - Opened by yuemenglong 7 months ago
- 1 comment
Labels: bug
#1584 - 微调seaco_paraformer模型单机多卡出现错误
Issue -
State: closed - Opened by zyjcsf 7 months ago
- 4 comments
Labels: question
#1577 - 中文语音中夹带少量英语时英语单词可能会缺失,
Issue -
State: closed - Opened by lzdk2003 8 months ago
- 1 comment
Labels: bug
#1575 - 如何屏蔽itn中带有货币形式的转换? TP99-->T九十九博茨瓦纳普拉
Issue -
State: closed - Opened by njzheng 8 months ago
Labels: question
#1573 - 'hdbscan' module not found; maybe use installed sklearn.cluster.HDBSCAN?
Issue -
State: closed - Opened by atomiechen 8 months ago
- 13 comments
Labels: question
#1572 - AttributeError: 'str' object has no attribute '__path__'
Issue -
State: closed - Opened by monkeycc 8 months ago
- 5 comments
Labels: question
#1570 - funasr 接口怎么对接rtsp音频流进行识别呢?
Issue -
State: closed - Opened by lonngxiang 8 months ago
Labels: question
#1564 - Failed to run whisper demo in this repo
Issue -
State: closed - Opened by BakerBunker 8 months ago
Labels: bug
#1560 - SeacoParaformer微调后效果变更差了,但是同样的数据,Paraformer-contextual微调后效果就能变好。
Issue -
State: closed - Opened by wjyfelicity 8 months ago
- 6 comments
Labels: question
#1550 - 短音频重复拼接为长音频后,声纹识别出现严重错误
Issue -
State: closed - Opened by jpyjpr 8 months ago
Labels: bug
#1546 - 调用funasr出现段错误
Issue -
State: closed - Opened by lonngxiang 8 months ago
- 1 comment
Labels: bug
#1545 - 在modelscope下載funasr的標點預測模型,在進行預測時出現錯from funasr import AutoModel,顯示AttributeError: partially initialized module 'torchaudio' has no attribute 'lib' (most likely due to a circular import)
Issue -
State: closed - Opened by alanshaoTT 8 months ago
- 1 comment
Labels: bug
#1544 - 什么时候出英文的实时识别
Issue -
State: closed - Opened by xiaojiang2022001 8 months ago
- 1 comment
Labels: question
#1543 - Code Error: Incorrect Path of Wav in VAD Model Example
Issue -
State: closed - Opened by nuaalixu 8 months ago
#1542 - self._handle = _dlopen(self._name, mode) OSError: [WinError 127] 找不到指定的程序
Issue -
State: closed - Opened by lonngxiang 8 months ago
- 1 comment
Labels: question
#1540 - 粤语识别模型推理出错,是否有长音频的模型
Issue -
State: closed - Opened by WuerLei 8 months ago
- 1 comment
Labels: bug
#1525 - fsmn vad 如何精确定位句尾?
Issue -
State: closed - Opened by bigmisspanda 8 months ago
- 1 comment
Labels: question
#1518 - Some warnings for training
Issue -
State: closed - Opened by dospeech 8 months ago
- 4 comments
Labels: question
#1516 - KeyError: 'speaker-diarization-inference is not in the pipelines registry group speaker-diarization.
Issue -
State: closed - Opened by san-kou7 8 months ago
- 2 comments
Labels: bug
#1512 - AliFsmnVad实例的GetSegments函数,返回的时间比实际时长多
Issue -
State: closed - Opened by greatbag 8 months ago
- 1 comment
Labels: bug
#1508 - question for told model
Issue -
State: closed - Opened by freshpearYoon 8 months ago
- 1 comment
Labels: question
#1501 - 支持多个音频批次推理吗
Issue -
State: closed - Opened by tuotuoshao 8 months ago
- 5 comments
Labels: question
#1500 - 请问SeACoParaformer热词语音识别模型现在支持lm了吗?
Issue -
State: closed - Opened by JVfisher 8 months ago
Labels: question
#1495 - compute_audio_cmvn.py 脚本运行时 Segmentation fault
Issue -
State: closed - Opened by hicliff 8 months ago
- 2 comments
Labels: bug
#1488 - Socket通信中pipeline执行speaker-verification任务出现choose a window size 400 that is [2, 0]报错
Issue -
State: closed - Opened by eugeneYz 8 months ago
- 1 comment
Labels: bug
#1487 - 请问VAD的max_end_silence_time参数怎么使用?
Issue -
State: closed - Opened by manbuheiniu 8 months ago
- 6 comments
Labels: question
#1487 - 请问VAD的max_end_silence_time参数怎么使用?
Issue -
State: closed - Opened by manbuheiniu 8 months ago
- 6 comments
Labels: question
#1471 - [src/libmpg123/parse.c:wetwork():1365] error: Giving up resync after 1024 bytes
Issue -
State: closed - Opened by qianqianque163645546464 8 months ago
- 1 comment
Labels: question
#1454 - runtime iOS Demo cannot compile successfully
Issue -
State: closed - Opened by summerHearts 8 months ago
- 5 comments
Labels: question
#1448 - KeyError: 'FunASRPipeline: KeyError("GenericFunASR: \'config\'")'
Issue -
State: closed - Opened by suzhaoxi 8 months ago
- 1 comment
Labels: question
#1436 - uniasr方言模型与vad 、punc模型一起使用时的bug
Issue -
State: closed - Opened by seanzhang-zhichen 8 months ago
- 5 comments
Labels: question
#1415 - funasr-runtime-sdk-online-cpu-0.1.9,异常
Issue -
State: closed - Opened by pony5551 8 months ago
- 2 comments
Labels: bug
#1397 - 发现一个长语音给funasr识别,单句单句输出,和两句两句输出,输出结果不一样
Issue -
State: closed - Opened by worldlanguage 9 months ago
Labels: bug
#1389 - 模型微调,加载训练数据后,报错: batch_size = input.size(0) AttributeError: 'str' object has no attribute 'size'
Issue -
State: closed - Opened by feifei-love 9 months ago
- 9 comments
Labels: question
#1384 - CT-Transformer标点模型的训练代码没有样例呢,求助!!!
Issue -
State: closed - Opened by lzl1456 9 months ago
- 4 comments
Labels: question
#1365 - python3 funasr_wss_client.py asyncio.exceptions.TimeoutError
Issue -
State: closed - Opened by willnufe 9 months ago
Labels: question
#1360 - 实时音视频转文字听写整句结果返回慢
Issue -
State: closed - Opened by L-Jim 9 months ago
- 5 comments
Labels: question
#1338 - python 版本的runtime使用websocket进行测试,长时间运行时会出现客户端无法向服务端发送消息的问题
Issue -
State: closed - Opened by Huanyongji 10 months ago
- 1 comment
Labels: bug
#1327 - TypeError: expected Tensor as element 1 in argument 0, but got str
Issue -
State: closed - Opened by OswaldoBornemann 10 months ago
- 14 comments
Labels: question
#1326 - automodel做后台服务,多次调用model.generate出现错
Issue -
State: closed - Opened by spritelw 10 months ago
- 4 comments
Labels: bug
#1326 - automodel做后台服务,多次调用model.generate出现错
Issue -
State: closed - Opened by spritelw 10 months ago
- 4 comments
Labels: bug
#1321 - If you run with error in funasr-1.0, please see ref to this issue
Issue -
State: closed - Opened by LauraGPT 10 months ago
Labels: question
#1313 - 请问一下怎么通过自己的数据集微调模型呀,之前版本的微调好像不见了
Issue -
State: closed - Opened by wang-ship-it 10 months ago
Labels: question
#1304 - FT2000 cpu deploy error
Issue -
State: closed - Opened by chinaguole 10 months ago
Labels: question
#1290 - recognizing in memory data recorded by pyaudio
Issue -
State: closed - Opened by lancemao 10 months ago
- 2 comments
Labels: question
#1285 - Not able to run diarization pipeline
Issue -
State: closed - Opened by pulkitmehtaworkmetacube 10 months ago
Labels: bug
#1253 - 希望能提供docker进程、FunASR进程崩溃后自动重启守护进程功能
Issue -
State: closed - Opened by leceshide 10 months ago
- 2 comments
Labels: question
#1240 - 出现:AutomaticSpeechRecognitionPipeline: cannot import name 'asr_inference_launch' from 'funasr.bin'
Issue -
State: closed - Opened by White-Friday 10 months ago
- 11 comments
#1240 - 出现:AutomaticSpeechRecognitionPipeline: cannot import name 'asr_inference_launch' from 'funasr.bin'
Issue -
State: closed - Opened by White-Friday 10 months ago
- 11 comments
#1233 - 中文符号判断的问题
Issue -
State: closed - Opened by laochen 10 months ago
#1230 - fintune问题
Issue -
State: closed - Opened by dospeech 10 months ago