speechcolab / leaderboard Goto Github PK

here:https://github.com/SpeechColab/Leaderboard/blob/master/models/tencent_api_zh/asr_api.py#L56
the base64Wav = base64.b64encode(data) should be base64Wav = str(base64.b64encode(data),encoding="utf-8")

It seems that the url for baidu to obtain the token has changed，it is not TOKEN_URL = 'http://openapi.baidu.com/oauth/2.0/token' now

哭了，请问下为啥我按照README来做还是有问题呀，命令及问题如下：

luody@cxh11:/mnt/database/luody/Leaderboard$ ops/pull -d SPEECHIO_ASR_ZH00001
2023-07-14 10:42:25,469 [INFO] Namespace(dataset='SPEECHIO_ASR_ZH00001', model=None)
2023-07-14 10:42:25,469 [ERROR] Please install oss via utils/install_aliyun_oss_client.sh

为啥会让我运行这个install_aliyun_oss_client.sh文件呀，可是我运行了又会让我验证

我在gigaspeech 上看到k2_gigaspeech模型的效果是10.40 / 10.51 ，如图所示：

而且我看k2 gigaspeech的测试结果 也是这个，如图所示：

但是为什么在Leaderboard上的结果对不上呢？如图所示：

Hi!
I want to run the wenetspeech model but find the error "decoder_main: command not found". How can I fix it? Thank you!

想问一下，我看model目录下面一些demo用到的接口是短语音识别（1分钟以内），如果我的测试语音超过了1分钟，是不是就要用录音文件转写的接口了

Hi,

Thanks for sharing your work, this is a great resource. I was wondering whether it would be possible to share the data behind leaderboard images that are in the readme. I'm new so if I these are already provided please point me to them instead.

Thank you.

如果我想上榜你们的排名榜，应该和谁联系

比如腾讯云的ASR大模型，头条的豆包ASR大模型等。

oss是一个html文件，直接用ops/pull里面的命令行运行会报错，这个怎么解决呢

root@multi-gpu-0:~/code/github/Leaderboard# ops/pull dataset SPEECHIO_ASR_ZH00006
2021-11-08 08:30:42,648 [INFO] Namespace(resource_id='SPEECHIO_ASR_ZH00006', resource_type='dataset')
Traceback (most recent call last):
File "ops/pull", line 42, in
src = remote_dataset_zoo[dataset_id]['url']
KeyError: 'SPEECHIO_ASR_ZH00006'

and the dataset/zoom.yaml file not contain the config info of SPEECHIO_ASR_ZH00006??

could you please provide instructions for downloading the test set?

I read your latest CER test results, in the test sets (ZH00001 ~ ZH00018), I would like to ask whether the test results refer to the combined test results of ZH00001 ~ ZH00018, or the average of separate tests

Thanks for the great the work!
I'm wondering whether the testing report about academic dataset (especially for the cloud model) is provided?
E.g. Microsoft API CER in aishell1 and aishell2

我已经配置了credential，为什么还是下载不了
Error: oss: service returned error: StatusCode=403, ErrorCode=AccessDenied, ErrorMessage="The bucket you access does not belong to you."

$ ./ops/pull -d SPEECHIO_ASR_ZH00000
2024-04-11 11:27:02,611 [INFO] Namespace(model=None, dataset='SPEECHIO_ASR_ZH00000')
2024-04-11 11:27:02,611 [ERROR] You need credential to use the leaderboard:
Please send email with title "oss.cfg" to [email protected], and paste replied content to credentials/aliyun_oss.cfg

一个月前就已经发了邮件，还没得到回复。

有哪位朋友可以分享一下这个文件吗？

Why does downloading according to specifications cause this issue

“
luody@cxh11:/mnt/database/luody/Leaderboard$ ops/pull dataset SPEECHIO_ASR_ZH00002
usage: pull [-h] [-m MODEL] [-d DATASET]
pull: error: unrecognized arguments: dataset SPEECHIO_ASR_ZH00002

”