FunASR

FunASR copied to clipboard

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

2

## ❓ Questions and Help 音频类模型也层出不穷，有通用添加该类模型支持的策略吗？相关文档？ https://github.com/MoonshotAI/Kimi-Audio

Kain-90

def export_rebuild_model(model, **kwargs): # self.device = kwargs.get("device") is_onnx = kwargs.get("type", "onnx") == "onnx" encoder_class = tables.encoder_classes.get(kwargs["encoder"] + "Export") model.encoder = encoder_class(model.encoder, onnx=is_onnx) predictor_class = tables.predictor_classes.get(kwargs["predictor"] + "Export") model.predictor = predictor_class(model.predictor,...

yuwei-star

pyinstaller6.11版本打包后运行报错

3

F:\gitwork\asr\runtime\python\websocket\dist>funasr_wss_server.exe model loading funasr version: 1.1.11. model_conf {'model': './models/iic/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch', 'model_revision': 'v2.0.4', 'ngpu': 1, 'ncpu': 4, 'device': 'cuda', 'disable_pbar': True, 'disable_log': True, 'disable_update': True} INFO:root:download models from model hub: ms Traceback...

SFidea

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

18

Notice: In order to resolve issues more efficiently, please raise issue following the template. （注意：为了更加高效率解决您遇到的问题，请按照模板提问，补充细节） ## 🐛 Bug Onnxruntime, 持续 2pass 模式，循环解码大概4小时，线程会卡死 ### To Reproduce onnxruntime c++ 最新版，持续解码大概4小时，会推理线程卡死，但是内存和cpu还是一直占着，通过 pstack 查看，是卡在...

locasxe

` from funasr import AutoModel from funasr.utils.postprocess_utils import rich_transcription_postprocess import funasr import time sensevoicesmall_model_dir = "./funasr_models/iic/SenseVoiceSmall" vad_model_dir = "./funasr_models/iic/speech_fsmn_vad_zh-cn-16k-common-pytorch" paraformer_model_dir = "./funasr_models/iic/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" punc_ct_model_dir = "./funasr_models/iic/punc_ct-transformer_cn-en-common-vocab471067-large" cam_model_dir = "./funasr_models/iic/speech_campplus_sv_zh-cn_16k-common" input_file =...

jinwater88

Notice: In order to resolve issues more efficiently, please raise issue following the template. （注意：为了更加高效率解决您遇到的问题，请按照模板提问，补充细节） ## ❓ Questions and Help #### What is your question? linux下torch环境太大了 想要导出复合模型 将复合模型导出后 SenseVoiceSmall只有一个model.onnx speech_fsmn_vad_zh-cn-16k-common-pytorch只有...

cstcood

funasr离线CPU版本新部署开始报ModuleNotFoundError: No module named 'zoneinfo'

1

I20250904 14:03:54.103415 418 funasr-wss-server.cpp:25] quantize : true I20250904 14:03:54.103431 418 funasr-wss-server.cpp:25] bladedisc : true I20250904 14:03:54.103444 418 funasr-wss-server.cpp:25] vad-dir : damo/speech_fsmn_vad_zh-cn-16k-common-onnx I20250904 14:03:54.103457 418 funasr-wss-server.cpp:25] vad-quant : true I20250904 14:03:54.103469...

zhilongquan

对于vad 模型，如何调整vad模型参数，能够过滤部分噪声？

1

如题，如何对于vad模型的参数进行调整，使得模型对于噪声不敏感；另外vad模型的参数有些疑惑，有没有关于vad参数的一些解释呀？

ldb1026

forward() is missing value for argument 'bias_embed'

3

启动命令： nohup bash run_server.sh \ --vad-dir /workspace/models/speech_fsmn_vad_zh-cn-16k-common-onnx \ --model-dir /workspace/models/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404 \ --punc-dir /workspace/models/punc_ct-transformer_cn-en-common-vocab471067-large-onnx\ --lm-dir /workspace/models/speech_ngram_lm_zh-cn-ai-wesp-fst \ --itn-dir /workspace/models/fst_itn_zh \ --certfile 0 \ --decoder-thread-num 20 \ --io-thread-num 10 \ --model-thread-num 16...

sf9ehf9fe

windows 11系统 软件版本: funasr-runtime-win-cpu-x64-v0.4.5 从日志中看,有两个错误: 1. type must be number, but is string 2. websocketpp.transport:7 (End of File)

GIEYang