espressif / esp-skainet

Espressif intelligent voice assistant
Other
587 stars 132 forks source link

esp-skainet\examples\chinese_tts总是自动重启 (AIS-1174) #96

Closed Henry586 closed 1 year ago

Henry586 commented 1 year ago

全新编译chinese_tts项目通过,但是总是自动重启,

仅仅使用idf.py menuconfig 设置 了一下 板子为Korvo-2,没有其他的改动: (Top) → Audio Media HAL Audio hardware board (ESP32-S3-Korvo-2) --->

启动后报告错误: `I (742) DRV8311: ES8311 in Slave mode I (752) gpio: GPIO[48]| InputEn: 0| OutputEn: 1| OpenDrain: 0| Pullup: 0| Pulldown: 0| Intr:0 voice_data paration size:3983360 init voice set:template ESP Chinese TTS v1.7 (Sep 22 2022 14:35:13, 1) 欢迎使用乐鑫语音合成 ESP-ROM:esp32s3-20210327 Build:Mar 27 2021 rst:0x8 (TG1WDT_SYS_RST),boot:0x8 (SPI_FAST_FLASH_BOOT) Saved PC:0x420023b1 0x420023b1: panic_handler at D:/esp/Espressif/frameworks/esp-idf-v4.4.4/components/esp_system/port/panic_handler.c:148 (discriminator 3)

SPIWP:0xee mode:DIO, clock div:1 load:0x3fce3808,len:0x17c4 load:0x403c9700,len:0xe78 load:0x403cc700,len:0x310c entry 0x403c9980`

请问这个怎么调试,是不是配置有问题呢? 所有的日志在附件中: tts_boot_error.txt

Henry586 commented 1 year ago

忘记写入语音库了,写入后就好了, 不过这个效果一言难尽啊

`d:\esp\esp-skainet\examples\chinese_tts>esptool.py --port com3 --baud 115200 --before default_reset --after hard_reset write_flash --flash_mode dio --flash_freq 40m --flash_size detect 0x410000 "esp_tts_voice_data_xiaoxin.dat" esptool.py v3.3.2 Serial port com3 Connecting... Failed to get PID of a device on com3, using standard reset sequence.

Detecting chip type... ESP32-S3 Chip is ESP32-S3 Features: WiFi, BLE Crystal is 40MHz MAC: 34:85:18:42:f9:c4 Uploading stub... Running stub... Stub running... Configuring flash size... Auto-detected Flash size: 16MB Flash will be erased from 0x00410000 to 0x0068bfff... Compressed 2602608 bytes to 2447971... Wrote 2602608 bytes (2447971 compressed) at 0x00410000 in 217.0 seconds (effective 96.0 kbit/s)... Hash of data verified.

Leaving... Hard resetting via RTS pin...`

Henry586 commented 1 year ago

close it

feizi commented 1 year ago

目前这个TTS是基于单词拼接的方法,词表大小限制了效果。
如果是针对数字播报应用,应该还是可以的,我们针对这个应用加了更多的音频单元,听上去会更自然一点。

Henry586 commented 1 year ago

谢谢大佬,回复这么神速!