2 Pull requests merged by 2 people

• fix bug for core dump in http, use libboost as parse

  #2509 merged May 8, 2025

• Update README.md

  #2487 merged Apr 22, 2025

1 Pull request opened by 1 person

• fix "can not find model issue when running libtorch runtime"

  #2504 opened May 6, 2025

8 Issues closed by 8 people

• 请教下训练环境的问题

  #2503 closed May 7, 2025

• 使用pyinstaller打包程序后运行exe文件出现TypeError: 'NoneType' object is not callable报错

  #2507 closed May 7, 2025

• 关于 onnxruntime 中的内存释放问题

  #2489 closed May 6, 2025

• 使用paraformer-zh为什么报错SeacoParaformer is not registered

  #2501 closed May 1, 2025

• 模型初始化1次，连续转写两个音频文件，第一个会走gpu，第二个不会走gpu。

  #2498 closed Apr 29, 2025

• 关于2pass websocket返回结果和直接使用在线模型推理结果不一样的问题

  #2493 closed Apr 27, 2025

• Paraformer ASR outputs uppercase garbage without whitespace (using lowercase + space char-level tokenizer)

  #2483 closed Apr 22, 2025

• 训练后没有生成cmvn.json和am.mvn

  #2480 closed Apr 16, 2025

25 Issues opened by 23 people

• Onnxruntime, 持续 2pass 模式，循环解码大概4小时，线程会卡死

  #2516 opened May 14, 2025

• python3 funasr_wss_client.py 命令行导出文件如何只返回文字

  #2515 opened May 14, 2025

• python版本的服务端不支持热词吗？

  #2514 opened May 14, 2025

• 增加 kimi-audio 的支持 / Add support for kimi-audio

  #2513 opened May 12, 2025

• funasr-runtime-sdk-online有支持GPU计划吗

  #2512 opened May 9, 2025

• 如何提升说话人和热词识别精度

  #2511 opened May 9, 2025

• 客户端无法返回文本

  #2510 opened May 9, 2025

• 什么时间会更新啊，最晚更新镜像时间还是2024/10/29

  #2506 opened May 6, 2025

• OSError: /usr/local/lib/python3.10/dist-packages/torchaudio/lib/libtorchaudio.so: undefined symbol: _ZN5torch8autograd15AutogradContextC1ERNS_6dynamo8autograd10PackedArgsE

  #2505 opened May 6, 2025

• 基于SenseVoiceSmall训练方言时报错误KeyError: 6491

  #2502 opened May 1, 2025

• Java GC overhead limit exceeded

  #2500 opened Apr 28, 2025

• 预训练模型转为ggml

  #2499 opened Apr 27, 2025

• FunASR hotword 是否支持设置热词权重

  #2497 opened Apr 27, 2025

• 安装时报错 提示cmake版本过低bug

  #2496 opened Apr 25, 2025

• RuntimeError: Failed to import transformers.modeling_utils because of the following error (look up to see its traceback): No module named 'torch.distributed.tensor'

  #2495 opened Apr 25, 2025

• 使用性能问题

  #2494 opened Apr 25, 2025

• BUG in EXPORT custom trained model.

  #2492 opened Apr 24, 2025

• 使用pyinstaller打包程序后，运行报错 FileNotFoundError

  #2491 opened Apr 23, 2025

• 自研了一套支持多路并发的asr日志服务

  #2490 opened Apr 22, 2025

• fun-onnx cuda-ep运行时快时慢

  #2488 opened Apr 22, 2025

• 如何传入定制热词训练seaco-paraformer

  #2486 opened Apr 21, 2025

• 训练SenseVoiceSmall模型出现train/loss值为负数

  #2485 opened Apr 18, 2025

• 请问可以跑在910B下吗

  #2484 opened Apr 18, 2025

• 请问FunASR 支持Intel的NPU运行吗？

  #2482 opened Apr 16, 2025

• 如何在一次调用服务同时实现角色识别、情感识别、方言等能力？

  #2481 opened Apr 16, 2025

26 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• funasr支持speech_campplus_speaker-diarization_common吗

  #2421 commented on Apr 15, 2025 • 0 new comments

• 基于paraformer & whisper 的离线转写网页，时间戳断句试听，生成字幕等

  #2059 commented on Apr 16, 2025 • 0 new comments

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

  #2441 commented on Apr 17, 2025 • 0 new comments

• 实时语音识别示例代码报错,显示返回的参数数量不对，求大神指点

  #2399 commented on Apr 21, 2025 • 0 new comments

• 逆文本正则化bug

  #2474 commented on Apr 22, 2025 • 0 new comments

• GPU版找不到文件model_blade.torchscript

  #2201 commented on Apr 23, 2025 • 0 new comments

• websocket python服务端使用热词无效

  #2471 commented on Apr 25, 2025 • 0 new comments

• 微调seaco_paraformer模型时自动结束

  #2467 commented on Apr 25, 2025 • 0 new comments

• 自建数据微调 seaco_paraformer 后，推理官方音频，预测器为什么只输出开始字符？

  #2342 commented on Apr 25, 2025 • 0 new comments

• 中文实时听写服务Docker镜像funasr-runtime-sdk-online-cpu-0.1.12存在内存泄露问题

  #2469 commented on Apr 27, 2025 • 0 new comments

• 部署没问题，无法调用

  #2461 commented on Apr 27, 2025 • 0 new comments

• 无法加载Qwen/Qwen-Audio

  #2478 commented on Apr 29, 2025 • 0 new comments

• funasr支持华为npu吗

  #1888 commented on Apr 30, 2025 • 0 new comments

• 为什么gpu版本没有流式回复了，识别速度很慢

  #2348 commented on Apr 30, 2025 • 0 new comments

• GPU版本 ws里面返回的is_final 一直是false状态

  #2141 commented on Apr 30, 2025 • 0 new comments

• 四川话识别哪个最优

  #2446 commented on May 3, 2025 • 0 new comments

• 微调模型，数据中未收录的单词并未影响预测结果

  #1996 commented on May 3, 2025 • 0 new comments

• funasr + Whisper语音识别-多语言-large-v3 + fsmn-vad + ct-punc-c + cam++ 报错 raise NotImplementedError("batch decoding is not implemented") NotImplementedError: batch decoding is not implemented

  #2273 commented on May 6, 2025 • 0 new comments

• 如何加入多个热词

  #2445 commented on May 6, 2025 • 0 new comments

• docker 使用sensevoice 模型不能加载LM模型

  #2430 commented on May 7, 2025 • 0 new comments

• html始终连接失败

  #1995 commented on May 8, 2025 • 0 new comments

• CHUNK_SIZE 和chunk_interval 是什么意思？

  #2338 commented on May 8, 2025 • 0 new comments

• 实时语音识别在online模式下，怎么获取2pass-offline的时间戳

  #2396 commented on May 8, 2025 • 0 new comments

• 离线识别http服务funasr-http-server收到POST请求时崩溃

  #1943 commented on May 8, 2025 • 0 new comments

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

  #2253 commented on May 13, 2025 • 0 new comments

• Fix crash in ASR tasks when lm is set to none in #2237

  #2263 commented on Apr 28, 2025 • 0 new comments