Using my own wav file to transcribe the results was wrong

[QingWind6]

I downloaded the wav audio from here: https://voiceage.com/wbsamples/in_mono/Sports.wav
And I used the following command: taskset -c 4-7 python -m useful_transformers.transcribe_wav Sports.wav
The following is the result of the operation, which is completely wrong with the content in the audio. May I ask why this is and how to solve it?

Thank you so much!

[marty1885]

Not sure if this is the cause. The input audio must be 16000Hz Mono. The audio you uploaded is 48000Hz. Try down sample it?

[QingWind6]

Let me have a try later. Thank you so much!

[shuffblow]

Hello，I want to ask you a question. when I used this command：taskset -c 4-7 python -m useful_transformers.transcribe_wav
It will show this information. The version of toolkit-lite2 and runtime have been updated 2.0.0b0.
The information:
E RKNN: [22:48:05.734] failed to submit!, op id: 0, op name: MatMul, flags: 0x5, task start: 0, task number: 16, run task counter: 0, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn)
E RKNN: [22:48:05.840] failed to submit!, op id: 0, op name: MatMul, flags: 0x5, task start: 0, task number: 16, run task counter: 0, int status: 0, please try updating to the latest version of the toolkit2 and runtime from: https://eyun.baidu.com/s/3eTDMk6Y (PWD: rknn)

[marty1885]

The version of toolkit-lite2 and runtime have been updated 2.0.0b0.

Which kernel and version of RKNPU are you using? I haven't move useful-transformers to v2.0.0 yet. For now it still runs on 1.6.0

[shuffblow]

The version of toolkit-lite2 and runtime have been updated 2.0.0b0.

Which kernel and version of RKNPU are you using? I haven't move useful-transformers to v2.0.0 yet. For now it still runs on 1.6.0

In this website https://github.com/airockchip/rknn-toolkit2/tree/master. The latest version v2.0.0.
I tried to run it on my orangepi5plus(RK3588).
Thank you very much! I will try v1.6.0 later

[marty1885]

@shuffblow 1.6.0 exists here. In a different repo
https://github.com/rockchip-linux/rknn-toolkit2/tree/master/rknpu2