example/speech_recognition/asr样例写了一个关键词识别程序，关键词识别后播放提升音失败的原因？

仿造example/speech_recognition/asr样例写了一个关键词识别程序，识别到关键词后，就播放提示音。目前关键词可以正确识别，就是播放提示音的时候就报错，报错信息如下，请各位帮忙看看。多谢！

RESULT: 你好 2.646811 4.470000s

E (7329) AUDIO_THREAD: Error creating task OUT_iis

E (7329) AUDIO_FORGE: C:/Users/y00355270/Desktop/esp-adf/components/esp-adf-libs/esp_codec/audio_forge.c:232 (sonic_open): Got NULL Pointer

E (7329) AUDIO_ELEMENT: [OUT_iis] audio_thread_create failed

E (7336) AUDIO_ELEMENT: [Audio_forge] AEL_STATUS_ERROR_OPEN,-1

W (7343) AUDIO_ELEMENT: [OUT_iis] Element has not create when AUDIO_ELEMENT_RESUME

W (7349) AUDIO_ELEMENT: [Audio_forge] audio_element_on_cmd_error,7

E (7357) ESP_AUDIO_TASK: Got element[OUT_iis] error:ESP_ERR_AUDIO_OPEN, stop pipeline[0]

E (7373) AUDIO_ELEMENT: [Audio_forge] Element already stopped

W (7380) AUDIO_ELEMENT: OUT-[DEC_mp3] AEL_IO_ABORT

W (7385) MP3_DECODER: output aborted -3

W (7335) FATFS_STREAM: No more data, ret:0

E (7394) AUDIO_ELEMENT: [IN_file] Element already stopped

E (7400) AUDIO_ELEMENT: [DEC_mp3] Element already stopped

E (7406) AUDIO_ELEMENT: [Audio_forge] Element already stopped

W (7413) ESP_AUDIO_TASK: Destroy the old pipeline, ELEMENT OCCUR ERROR

W (7420) AUDIO_ELEMENT: [OUT_iis] Element has not create when AUDIO_ELEMENT_TERMINATE, tick:300

W (7429) ESP_AUDIO_TASK: The old pipeline destroyed, ELEMENT OCCUR ERROR

E (12364) task_wdt: Task watchdog got triggered. The following tasks did not reset the watchdog in time:

E (12364) task_wdt:  - IDLE (CPU 0)

E (12364) task_wdt: Tasks currently running:

E (12364) task_wdt: CPU 0: main

E (12364) task_wdt: CPU 1: IDLE

E (12364) task_wdt: Print CPU 0 (current core) backtrace