[TTS] mutiple pause mark in chinese sentences made TTSExecutor killed

For support and discussions, please use our Discourse forums.

If you've found a bug then please create an issue with the following information:

Describe the bug

Testing txts: 本门晚一辈的，男的是：诸葛警我、岳雯、严人英、金蝉、石生、庄易、林寒、白侠孙南、石奇、赵燕儿、杨鲤、龙力子、七星手施林、神眼邱林、苦孩儿司徒平、铁沙弥悟修、黑孩儿尉迟火、云中鹤周淳、易家双矮易鼎和易震、南海双童甄艮和甄兑、独霸川东李震川、灵和居士徐祥鹅、周云从、商风子、章虎儿、张琪、黄玄极等；女的是：齐灵云和霞儿姊妹、李英琼、余英男、秦紫玲和寒萼姊妹、墨凤凰申若兰、女神童朱文、女殃神郑八姑、周轻云、女空空吴文琪、红娘子余莹姑、女神婴易静、廉红药、凌云凤、裘芷仙、章南姑、郁芳蘅、李文行、万珍、云紫绢、陆蓉波、金萍、赵铁娘，以及由金姥姥罗紫烟转引到本门的女飞熊吴玫、女大鹏崔绮、美仙娃向芳淑等。

To Reproduce Steps to reproduce the behavior:

1. python code as below:

   try: # tts_executor = TTSExecutor() wav_file = tts_executor(
   text=text,
   output=outputpath,
   am=model_cfg['am'],
   am_config=model_cfg['am_config'],
   am_ckpt=model_cfg['am_ckpt'],
   am_stat=model_cfg['am_stat'],
   spk_id=model_cfg['spk_id'],
   phones_dict=model_cfg['phones_dict'],
   tones_dict=model_cfg['tones_dict'],
   speaker_dict=model_cfg['speaker_dict'],
   voc=model_cfg['voc'],
   voc_config=model_cfg['voc_config'],
   voc_ckpt=model_cfg['voc_ckpt'],
   voc_stat=model_cfg['voc_stat'],
   lang=model_cfg['lang'],
   device=paddle.get_device()) print("@an mixTxt2Wav(),tts_executor",sys.getsizeof(wav_file)) except Exception as e : print("Err Text:",text) os.system("echo '"+text+"' >> ./err.log") print(e) exit(1) print("output:",mp3Filepath)

Expected behavior expected wav file generated, but it failed as "killed"

Screenshots

Environment (please complete the following information):

• OS: [e.g. Ubuntu]
• GCC/G++ Version [e.g. 8.3]
• Python Version [e.g. 3.7]
• PaddlePaddle Version [e.g. 2.0.0]
• Model Version [e.g. 2.0.0]
• GPU/DRIVER Informationo [e.g. Tesla V100-SXM2-32GB/440.64.00]
• CUDA/CUDNN Version [e.g. cuda-10.2]
• MKL Version
• TensorRT Version

Additional context Add any other context about the problem here.