Still getting "utils/internal/download_gigaspeech_with_pyspeechcolab.sh: This recipe needs the package speechcolab installed." message

[makwadajp]

I ran the following command: sudo bash utils/download_gigaspeech.sh /mnt/z/GigaSpeech

However I ran into the following error when executing:

utils/download_gigaspeech.sh: Downloading with PySpeechColab...
utils/internal/download_gigaspeech_with_pyspeechcolab.sh: This recipe needs the package speechcolab installed.
   To install:
       pip install speechcolab

So I have pip installed the speechcolab package. When I reran the above command, I still get the same error message. I have checked that pip list shows speechcolab 0.0.6a0, so how should I proceed from here?

[makwadajp]

I reinstalled Ubuntu from scratch on WSL (On Windows Command Prompt: wsl -l -v → VERSION 1). THIS WORKED. EDIT: So now, I do not know what was wrong originally. But if it helps anybody then here is what I did: 　 　 I also figured out what will result in This recipe needs the package speechcolab installed. error message. 　 　 Please see here for what the cause was.

1. Install Ubuntu 22.04 LTS from Microsoft Store
2. On WSL, execute sudo apt update && sudo apt -y upgrade
3. Execute sudo apt install python3-pip
4. Download appropriate Anaconda (in my case 64-Bit (x86) Installer (581 MB) so wget https://repo.anaconda.com/archive/Anaconda3-2021.11-Linux-x86_64.sh)
5. Execute bash ./Anaconda3-2021.11-Linux-x86_64.sh
6. Exit WSL (You need to say yes to Do you wish the installer to initialize Anaconda3 by running conda init? [yes|no] during the installation of Anaconda at step 5 and exit WSL. Otherwise you will encounter conda: command not found at step 7)
7. Restart WSL and execute conda create --name GigaSpeech
8. Execute conda activate GigaSpeech
9. Execute pip install speechcolab sudo pip install speechcolab
10. Execute sudo mkdir /mnt/z (create a directory for mounting a network drive to copy 1.2TB of data)
11. Execute sudo mount -t drvfs z: /mnt/z (this step needs to be called everytime the computer is restarted or even possibly when WSL is restarted)
12. Execute the instructions sent from gigaspeech@speechcolab.org ... sudo bash utils/download_gigaspeech.sh /mnt/z/GigaSpeech

For people installing GigaSpeech using WSL, here are other information:

1. Regarding WSL installation, I just followed Microsoft's Install WSL instruction page. I've installed: Ubuntu 22.04 LTS (GNU/Linux 4.4.0-19041-Microsoft x86_64)
2. If you receive the following error message when running instruction from gigaspeech@speechcolab.org (sudo git clone https://github.com/SpeechColab/GigaSpeech.git) then execute from a locally mounted drive (e.g. /mnt/c/) instead of a network drive (in my case /mnt/z/).

fatal: Unable to create temporary file '/mnt/z/GigaSpeech/.git/objects/pack/tmp_pack_XXXXXX': Permission denied

[makwadajp]

I re-encountered the same This recipe needs the package speechcolab installed. problem on my cleanly installed setup. However this stackoverflow: Unable to import a module that is definitely installed gave me a hint for the solution of the problem.

It works when both of the following commands were executed with `sudo':

• sudo pip install speechcolab
• sudo bash utils/download_gigaspeech.sh /mnt/z/GigaSpeech

As an error you will encounter if both are not called with sudo:

• calling bash utils/download_gigaspeech.sh /mnt/z/GigaSpeech without sudo will result in the following error:

PermissionError: [Errno 13] Permission denied: '/mnt/z/GigaSpeech/TERMS_OF_ACCESS'

• when sudo bash utils/download_gigaspeech.sh /mnt/z/GigaSpeech is called, calling pip install speechcolab without sudo will result in the following error (the error which frustrated me in the first place):

This recipe needs the package speechcolab installed.