多人区分说话报错

[deveropers]

报错：string index out of range

File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/gradio/queueing.py", line 625, in process_events
response = await route_utils.call_process_api(
ile "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/gradio/route_utils.py", line 322, in call_process_ap output - await app.get_blocks(.process_api(00, 7.33it/s]
File "/root/mini conda3/envs/FunClip/lib/python3.10/site-packages/gradio/blocks.py"
" Line 2103, in process_api
result = await self.call_function(
File "/root/mini conda3/envs/FunClip/lib/python3.10/site-packages/gradio/blocks.py", line 1650, in call_function
prediction = await anyio.to_thread.run_sync( # type: ignore
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync return await get_async_backend.run_sync_in_worker_thread(
Le "/root/miniconda3/envs/FunClip/Lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2461, in run_sync_in_worker_thread
return await future
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/anyio/_backends/_asyncio.py"
', line 962, in run
result = context. run(func, *args)
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/gradio/utils.py", line 890, in wrapper Fite "ogesrf/Func-ip/funclip/launch.py", ine 88, in mix recogspeaker
res_text, res_srt, audio_state = audio_recog(
ile "/opt/srt/FunClip/funclip/launch.py", line 50,
, line 50, in audio_recog
return audio_clipper.recog(audio_input, sd_switch, None, hotwords, output_dir=output_dir)
博（weichuanbo）
233-训练
File "/opt/srt/FunClip/funclip/videoclipper.py", line 47, in recog
春
rec_result = self. funasr_model.generate(data,
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/funasr/auto/auto_model.py", line 306, in generate return self.inference_with_vad(input, input_len=input_len, **cfg)
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/funasr/auto/auto_model.py", line 626, in inference_with_vad
sentence_list = timestamp_sentence(
File "/root/miniconda3/envs/FunClip/lib/python3.10/site-packages/funasr/utils/timestamp_tools.py"
', line 163, in timestamp_sentence
sentence_text_seg[:-1] if sentence_text_seg[-1] = " " else sentence_text_seg
IndexError: string index out of range

[deveropers]

经社区群确认为 funasr 1.2.5 版本 bug，暂时解决方案，手动降级 funasr 版本：pip install funasr==1.2.4

[Patrick-Ni]

1.2.6 依然有这个bug

[LauraGPT]

重新打开。已有用户明确反馈 FunASR 1.2.6 仍会在多人区分说话时报错，之后没有维护者复测或用户确认，因此原关闭状态不成立。

请用当前 FunClip main + funasr==1.4.7 复测，并提供完整 traceback、启动参数、是否启用 spk_model 以及输入音频时长/声道数。我们会按当前版本复现，而不是继续沿用“降级到 1.2.4”作为结论。

[LauraGPT]

补充可复测结论：旧堆栈中 timestamp_sentence() 对空 sentence_text_seg 做尾字符索引的路径，已经在 FunASR main@a63ededd85a1b42ee7ce892eb44e48cdc234d922 中加入空值保护（源自 #2429）。我在该 exact main 的隔离工作树运行了 tests/test_timestamp_tools.py、tests/test_paraformer_timestamp_contract.py 和 tests/test_paraformer_timestamp_padding.py，共 9 passed。

请升级到 funasr==1.4.13 与 FunClip v2.2.1 后，用原多人音频复测，不再建议降级。若仍报错，请贴出完整 traceback 和 python -m pip show funasr funclip 输出；本 issue 保持打开，等待原环境确认。