Issues · modelscope/FunASR · GitHub

Labels Milestones New issue

英文在线识别

#2849

· holylong opened

on Apr 30, 2026

[runtime/onnxruntime] Paraformer::Forward heap-buffer-overflow: encoder_out_lens reads ONNX int32 tensor as int64_t

#2848

· wangxiuwen opened

on Apr 29, 2026

[runtime/onnxruntime] AudioFrame copy constructor missing data/global_start/global_end (rule-of-three violation)

#2847

· wangxiuwen opened

on Apr 29, 2026

[runtime/onnxruntime] Audio::FfmpegLoad(buf) early-return paths leak n_file_len bytes via avio_context_free

#2846

· wangxiuwen opened

on Apr 29, 2026

[runtime/onnxruntime] Memory leak: new[] without delete[] in FunTpassInferBuffer second loop

#2845

· wangxiuwen opened

on Apr 29, 2026

fun-asr-nano可以支持时间戳返回和说话人分离吗？

#2844

· dimplesT2 opened

on Apr 28, 2026

使用java对接FunASR离线文件转写服务怎样支持说话人识别？

#2843

· jiangdonglin1 opened

on Apr 14, 2026

Automodel.generate()中的punc_model 空参及None bug

#2839

· noforlo opened

on Mar 27, 2026

pip包什么时候更新？

#2838

· sandsc opened

on Mar 24, 2026

Wrong wav_format: mp4

#2837

· funny-clivia opened

on Mar 23, 2026

Fun-ASR-Nano 有支持实时识别的计划吗

#2836

· ljh1994 opened

on Mar 20, 2026

Go调用websocket接口离线转写会提前中断websocket连接

#2834

· ibarryyan opened

on Mar 17, 2026