assemblyai / assemblyai-cli Goto Github PK

Please describe your feature request

Add output file flag to write the transcribe text. Currently assemblyai write the text to stdout.

This can be done with bash redirect

assemblyai transcribe audio.mp4 > file.txt

I think it is good to support this out of box

assemblyai transcribe audio.mp4 --output file.txt

The flag name for the output can be -f,--output-file or any other.

Checks need to be performed if the flag is given before transcribing. Like checking the file doesn't exist, also the folder exists in case of output some/path/to/file.txt. Check some/path/to/ exists, but file.txt doesn't exists in the folder. Check permission to write/create file.txt before transcribing.

Describe the bug
install script not working

To Reproduce
Steps to reproduce the behavior:

1. git clone https://github.com/AssemblyAI/assemblyai-cli.git
2. /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/AssemblyAI/assemblyai-cli/main/install.sh)"
3. Using the OSS distribution...
   Downloading AssemblyAI CLI v1.14.2...
   Installing AssemblyAI to /home/dirk/.assemblyai-cli
   Welcome to the AssemblyAI CLI!
   Please start by running assemblyai config [token]
   /bin/bash: Zeile 71: zsh: Befehl nicht gefunden

Expected behavior
Installation should work

Desktop (please complete the following information):

• OS: Ubuntu
• CLI Version v1.14.2

Describe the bug
I installed CLI using homebrew , then I try to do authentication using command

assemblyai config [mytoken]

but CTL return something run :

Something went wrong . Please try it again .

Expected behavior
I need to fix this ,and authentic successfully

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: [e.g. macOS]
• CLI Version v1.17.1

Additional context
Add any other context about the problem here.

Please describe your feature request
I want to be able to access the subtitle endpoint for after transcripts are finished processing so that it would be possible to download .srt files or pipe the text output into another API like Mux which is a video API.

Describe alternatives you've considered
I've used this with CURL requests and python, but prefer the assemblyai-cli.

Additional context
The Mux docs show how to add subtitles when creating an asset or you can also add them later. But you need to pass a URL to the .srt file, which means assemblyai-cli should have the option to download the .srt file as well as provide the text output.

Thanks in advance for your consideration.

Description
Installation fails on Windows when executing the second command
iex "$(curl https://raw.githubusercontent.com/AssemblyAI/assemblyai-cli/main/install.ps1)".

Output:

- Unexpected token '"-i"' in expression or statement.

To Reproduce
Run the following commands in PowerShell

Set-ExecutionPolicy RemoteSigned -Scope CurrentUser
iex "$(curl https://raw.githubusercontent.com/AssemblyAI/assemblyai-cli/main/install.ps1)"
New-Alias -Name assemblyai -Value C:\\'./Program Files\'/AssemblyAI/assemblyai.exe

Desktop:

• OS: Windows 10
• CLI Version: PowerShell 7

Please describe your feature request
When I fetch the transcript, say, when I'm polling, the response does not include the time the transcription finished.

However, the completion time is exposed when I export transcript jobs via CSV.

This makes it troublesome to synchronize data on my side. For my monitoring, it's helpful to have logs for when things started, failed, etc.

Ideally, the CSV export data and response should align so I can better reconcile the job data:

• uploaded timestamp
• transcription job start timestamp
• transcription job completed timestamp
• transcription job failed timestamp

Describe alternatives you've considered
I'm just regularly importing the CSV to capture the missing completion times and doing my best to reconcile.

Please forgive me if I am missing some other obvious way of processing the json.

Describe the bug
Using the -j flag on the get command for a transcript adds a line "Transcribing file with id rxzk61qto3-0573-48f4-824e-2a161985ef53" which causes jq (JSON parser) to return an error.

IMO, this line is unnecessary or there should be an option to remove it.

To Reproduce
Steps to reproduce the behavior:

1. Install jq to parse the JSON
2. Run assemblyai get {transcript ID} -j > id.json
3. jq . id.json
4. parse error: Invalid numeric literal at line 1, column 13

Expected behavior
jq should parse the transcript's json file without issue.

Desktop (please complete the following information):

• OS: macOS
• CLI Version v1.10.2

Additional context
If there's another workflow that I'm unaware of to process the output (whether it uses json or not), please let me know!

Describe the bug
When attempting to install the CLI on a raspberry pi, it fails without any errors

To Reproduce

1. SSH into the raspberry pi
2. Execute /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/AssemblyAI/assemblyai-cli/main/install.sh)"

Expected behavior
Installs the CLI

Experienced behavior
Prints

Using the OSS distribution...
Downloading AssemblyAI CLI v1.17.4...

No errors or other information is printed.

Desktop (please complete the following information):

$ cat /etc/os-release
PRETTY_NAME="Raspbian GNU/Linux 11 (bullseye)"
NAME="Raspbian GNU/Linux"
VERSION_ID="11"
VERSION="11 (bullseye)"
VERSION_CODENAME=bullseye
ID=raspbian
ID_LIKE=debian

Additional context
n.a.

I want test the flag "--custom_spelling ./custom_spelling.json", but I got an Error:
"Error parsing custom spelling file"

The json file just contains the example from here:
https://www.assemblyai.com/docs/core-transcription#custom-spelling

What am I doing wrong?

When transcribing a video, cli has crashed with a traceback; doing assemblyai get crashed in exactly the same spot:

$ assemblyai get rptv3lsl5w-8310-4944-a086-d49f5e5a26b5                                                                                         2
Transcribing file with id rptv3lsl5w-8310-4944-a086-d49f5e5a26b5
Transcript
panic: runtime error: index out of range [1] with length 1

goroutine 1 [running]:
github.com/AssemblyAI/assemblyai-cli/utils.GetSentenceTimestamps({0x140006be000?, 0x29c, 0x101222b01?}, {0x14001214000, 0x7507, 0x100f36908?})
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/utils.go:272 +0x634
github.com/AssemblyAI/assemblyai-cli/utils.textPrintFormatted({0x1400033c000, 0x26686}, {0x14001214000, 0x7507, 0xa005})
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:339 +0x114
github.com/AssemblyAI/assemblyai-cli/utils.getFormattedOutput({0x14000332ae0, 0x14000405788, 0x0, 0x0, 0x14000332b10, 0x140004057ac, 0x1400040579a, 0x0, {0x0, 0x0}, ...}, ...)
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:298 +0xcc
github.com/AssemblyAI/assemblyai-cli/utils.PollTranscription({0x16ef9f72f, 0x26}, {0x0?, 0x0?})
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:280 +0x2d0
github.com/AssemblyAI/assemblyai-cli/cmd.glob..func2(0x101679460?, {0x140000694b0?, 0x1?, 0x1?})
	/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/get.go:45 +0x124
github.com/spf13/cobra.(*Command).execute(0x101679460, {0x14000069490, 0x1, 0x1})
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:876 +0x4b8
github.com/spf13/cobra.(*Command).ExecuteC(0x1016796e0)
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:990 +0x354
github.com/spf13/cobra.(*Command).Execute(...)
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:918
github.com/AssemblyAI/assemblyai-cli/cmd.Execute()
	/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/root.go:42 +0xa8
main.main()
	/home/runner/work/assemblyai-cli/assemblyai-cli/main.go:11 +0x1c

You've got transcription ID right in there, so please investigate.

macOS 13.2, AssemblyAI CLI v1.14.2.

Installed latest Xcode CLI and assenblyai
with

brew tap assemblyai/assemblyai
brew install assemblyai

Unfortunately it doesn't work on my mac.
Message :

 ~ % assemblyai transcribe https://youtu.be/0wvBu014E5o -auto_highlights - entity_detection
Transcribing a youtube video....
Something went wrong. Please try again. 

I try several youtube links.

local mp3 file transcripts have works

hi there i installed you software but it creates an unknown user -- > [\e]0;\u@\h: \w\a]\u@\h:\w$ when i finished installing it and the software only works from with in this user. from the normal user it does not work and has this error

root@OpenWrt:~# assemblyai transcribe "/root/recording.wav"
Usage: assemblyai [OPTIONS]
Try 'assemblyai --help' for help.

Error: Got unexpected extra arguments (transcribe /root/recording.wav)

currently logged in from ssh so once I log out of the unknown user there be no way to log back in

as a foot note I had issue with the install script it uses rev and i had no rev on my device so I replaces rev with sed

test -z "$VERSION" && VERSION="$(curl -sfL -o /dev/null -w %{url_effective} "$RELEASES_URL/latest" |
                sed $'s/./&\\\n/g' | sed -ne $'x;H;${x;s/\\n//g;p;}' |
                cut -f1 -d'/'|
                sed $'s/./&\\\n/g' | sed -ne $'x;H;${x;s/\\n//g;p;}')"

it is install on 86_64 type device

Describe the bug
I get an error RuntimeError: Transcription failed: File does not appear to contain audio. File type is text/html (HTML document, UTF-8 Unicode text, with very long lines). when I send a youtube link

To Reproduce

Steps to reproduce the behavior:
import requests
import json
import time

base_url = "https://api.assemblyai.com/v2"
headers = {
    "authorization": Auth
}


data = {
    "audio_url": "https://www.youtube.com/watch?v=Abo19IokvKs&ab_channel=JimmyKimmelLive"
}

url = base_url + "/transcript"
response = requests.post(url, json=data, headers=headers)

transcript_id = response.json()['id']
polling_endpoint = f"https://api.assemblyai.com/v2/transcript/{transcript_id}"

while True:
    transcription_result = requests.get(polling_endpoint, headers=headers).json()

    if transcription_result['status'] == 'completed':
        print(transcription_result['text'])
        break

    elif transcription_result['status'] == 'error':
        raise RuntimeError(f"Transcription failed: {transcription_result['error']}")

    else:
        time.sleep(3)

Expected behavior
Transcription of video audio

Screenshots

Desktop (please complete the following information):

• OS: [Windows]
• Using python code

Additional context

I am trying to run the streamlit app and am getting this issue. It seems to be an issue with my microphone my Mac. My microphone is enabled but I am still getting the issue when running the app. Why is this happening?

OSError: [Errno -9999] Unanticipated host error
Traceback:
File "/Users/camabernethy/opt/anaconda3/lib/python3.9/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 552, in _run_script
exec(code, module.dict)
File "/Users/camabernethy/Real-time-transcription-from-microphone/sr_in_streamlit.py", line 22, in
stream = p.open(
File "/Users/camabernethy/opt/anaconda3/lib/python3.9/site-packages/pyaudio/init.py", line 639, in open
stream = PyAudio.Stream(self, *args, **kwargs)
File "/Users/camabernethy/opt/anaconda3/lib/python3.9/site-packages/pyaudio/init.py", line 441, in init
self._stream = pa.open(**arguments)

Expected behavior
Ability to record.

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: [e.g. macOS]
• CLI Version [e.g. 0.22]

In YoutubeDownload, the global variable Filename is overwritten, making it difficult to import and use these functions in external packages:

My use case is that I'm trying to make use of the AssemblyAI API, and I am re-using some of your functions in my own go program.

Describe the bug
Providing a Discord webhook URL as the string value to the the --webhook_url parameter to the transcribe command

To Reproduce
Steps to reproduce the behavior:

1. Create a Discord webhook
2. Test the webhook using Postman, confirming that it works as expected
3. Use the same URL as the value for --webhook_url
4. Nothing happens in Discord

Expected behavior
I should see something in Discord

Desktop (please complete the following information):

• OS: macOS
• CLI Version 1.17.3

Please describe your feature request
At the moment you can use --srt in the CLI to generate an srt file at <id>.srt. However, I'd like to specify the path the srt file would be stored at.

Describe alternatives you've considered
Currently I'm doing the following

video_path=...
srt_path="${video_path%.*}.srt"
tmp_path=$(assemblyai transcribe <file> --srt | tail -n 1 | awk '{print $NF}')
mv "${tmp_path}" "${srt_path}"

Additional context
n.a.

Hi,

I wanted to install the assemblyai-cli on Linux:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/AssemblyAI/assemblyai-cli/main/install.sh)"

=>

Using the OSS distribution...
Downloading AssemblyAI CLI v1.9.1...

No more output. 🙂

Environment

OS: Ubuntu 22.04

Troubleshooting

It looks like the script is trying to download
$RELEASES_URL/download/$VERSION/${FILE_BASENAME}_${VERSION:1}_$(uname -s)_$(uname -m).tar.gz

https://github.com/AssemblyAI/assemblyai-cli/blob/main/install.sh#L25

In my env:

uname -s

=>
Linux

uname -m

=>

x86_64

But there's no version to download with the suffix "linux_x86_64.tar.gz" among the releases

Describe the bug
Access to AssemblyAI is denied

To Reproduce
Steps to reproduce the behavior:

1. Following instruction on installing via powershell

2. Copy pasted the command to install assemblyai via powershell

3. See error
   New-Item : Access to the path 'AssemblyAI' is denied.
   At line:34 char:3

• New-Item -ItemType Directory -Path $INSTALL_DIR

• + CategoryInfo          : PermissionDenied: (C:\Program Files\AssemblyAI:String) [New-Item], UnauthorizedAccessException
  + FullyQualifiedErrorId : CreateDirectoryUnauthorizedAccessError,Microsoft.PowerShell.Commands.NewItemCommand
  
  

Copy-Item : Could not find a part of the path 'C:\Program Files\AssemblyAI\assemblyai.exe'.
At line:37 char:1

• Copy-Item $BINARY $TARGET -Force

•   + CategoryInfo          : NotSpecified: (:) [Copy-Item], DirectoryNotFoundException
    + FullyQualifiedErrorId : System.IO.DirectoryNotFoundException,Microsoft.PowerShell.Commands.CopyItemCommand
  
  

Start-Process : This command cannot be run due to the error: The system cannot find the file specified.
At line:39 char:1

• Start-Process -FilePath $TARGET -ArgumentList "welcome -i -o=windows ...

•   + CategoryInfo          : InvalidOperation: (:) [Start-Process], InvalidOperationException
    + FullyQualifiedErrorId : InvalidOperationException,Microsoft.PowerShell.Commands.StartProcessCommand
  
  

Expected behavior
It should install

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: WIndows 11
• CLI Version Latest version according to my written date

Additional context
Add any other context about the problem here.

Please consider enabling subscribers to use PayPal of Google Pay for payment.

Please describe your feature request
Add ability to list all transcripts. This would make it easier to follow some guides where the transcript ID's are required, such as LeMUR related guides.

Please describe your feature request
Word boost and custom vocabulary are two really important features of AssemblyAI (and other transcription services) but are for some reason not included in assemblyai-cli. Having these features would make scripting it WAY easier in my opinion.

Describe alternatives you've considered
I tried using curl requests, but find it to be too cumbersome compared to the ease of assemblyai-cli.

Additional context
I have tweaked yt-dlp extensively to be able to download videos and embed metadata, so being able to transcribe them via the cli would improve the ability to catalog clips effectively. All the other features are great and I use them, but many niches have lots of custom vocabulary.

I'm recieving this output whenever I run the given command ( assemblyai transcribe "youtube url" ) :-

'''
panic: runtime error: invalid memory address or nil pointer dereference
[signal 0xc0000005 code=0x0 addr=0x0 pc=0x95e8db]

goroutine 1 [running]:
github.com/AssemblyAI/assemblyai-cli/utils.YoutubeDownload({0xc00001e440?, 0xc0003755f0?})
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/youtube.go:56 +0x2bb
github.com/AssemblyAI/assemblyai-cli/utils.Transcribe({{0xc00001e420, 0x2b}, 0x0, 0x0, 0x0, 0x0, {0x0, 0x0, 0x0}, 0x0, ...}, ...)
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:78 +0x1bf
github.com/AssemblyAI/assemblyai-cli/cmd.glob..func4(0xdf9bc0?, {0xc000301710?, 0x1?, 0x1?})
/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/transcribe.go:249 +0xcf0
github.com/spf13/cobra.(*Command).execute(0xdf9bc0, {0xc0003016f0, 0x1, 0x1})
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:876 +0x67b
github.com/spf13/cobra.(*Command).ExecuteC(0xdf9940)
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:990 +0x3bd
github.com/spf13/cobra.(*Command).Execute(...)
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:918
github.com/AssemblyAI/assemblyai-cli/cmd.Execute()
/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/root.go:42 +0x85
main.main()
/home/runner/work/assemblyai-cli/assemblyai-cli/main.go:11 +0x17
'''

I don't know, if I can post it here.
It's not a bug, I think.

But I got the following error message. What does it mean? It's only when i try one specific mp3-file.
Other files worked well without any issue.

Can somebody help me?

`panic: runtime error: index out of range [7192] with length 7192

goroutine 1 [running]:
github.com/AssemblyAI/assemblyai-cli/utils.GetSentenceTimestamps({0xc00031c000?, 0x111, 0x1051601?}, {0xc00086c000, 0x1c18, 0xc0001fb4d8?})
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/utils.go:274 +0x71a
github.com/AssemblyAI/assemblyai-cli/utils.textPrintFormatted({0xc0002d4000, 0xaaa6}, {0xc00086c000, 0x1c18, 0x1f9c})
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:381 +0x13c
github.com/AssemblyAI/assemblyai-cli/utils.getFormattedOutput({0xc000088630, 0xc0002d0f78, 0x0, 0x0, 0xc000088660, 0xc0002d0f9c, 0xc0002d0f89, 0x0, {0x0, 0x0}, ...}, ...)
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:337 +0xd3
github.com/AssemblyAI/assemblyai-cli/utils.PollTranscription({0xc0000c8b70, 0x26}, {0x80?, 0x5?, 0x0?})
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:319 +0x428
github.com/AssemblyAI/assemblyai-cli/utils.Transcribe({{0xc000026190, 0x46}, 0x0, 0x0, 0x0, 0x0, {0x0, 0x0, 0x0}, 0x0, ...}, ...)
/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:154 +0x672
github.com/AssemblyAI/assemblyai-cli/cmd.glob..func4(0x1058aa0?, {0xc00002c6f0?, 0x3?, 0x3?})
/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/transcribe.go:261 +0xdd0
github.com/spf13/cobra.(*Command).execute(0x1058aa0, {0xc00002c6c0, 0x3, 0x3})
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:876 +0x67b
github.com/spf13/cobra.(*Command).ExecuteC(0x1058820)
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:990 +0x3bd
github.com/spf13/cobra.(*Command).Execute(...)
/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:918
github.com/AssemblyAI/assemblyai-cli/cmd.Execute()
/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/root.go:42 +0x85
main.main()
/home/runner/work/assemblyai-cli/assemblyai-cli/main.go:11 +0x17`

Describe the bug
I tried to run the following command:

assemblyai transcribe https://www.youtube.com/watch\?v\=xUqN2kj6Tt0 --auto_highlights --entity_detection

to transcribe a brazilian portuguese video, but got the output:

Transcribing YouTube video...
Downloading video:  5.77 MB / 5.77 MB  100.00% 1s                                                         
Uploading file to our servers:  5.77 MB / 5.77 MB  100.00% 20s                                            
Transcribing file with id r2saqo91d8-0cc5-4b7e-9256-1d19edcd7b35
Transcript


Highlights
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x0 pc=0x10450eed4]

goroutine 1 [running]:
github.com/AssemblyAI/assemblyai-cli/utils.getFormattedOutput({0x140000e3eb0, 0x140000275e0, 0x0, 0x0, 0x140000e3ee0, 0x140000275f7, 0x140000275f0, 0x0, {0x0, 0x0}, ...}, ...)
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:306 +0x194
github.com/AssemblyAI/assemblyai-cli/utils.PollTranscription({0x140000ce4e0, 0x26}, {0x80?, 0x5?})
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:280 +0x2d0
github.com/AssemblyAI/assemblyai-cli/utils.Transcribe({{0x1400002bd10, 0x46}, 0x0, 0x1, 0x0, 0x0, {0x0, 0x0, 0x0}, 0x0, ...}, ...)
	/home/runner/work/assemblyai-cli/assemblyai-cli/utils/transcribe.go:144 +0x588
github.com/AssemblyAI/assemblyai-cli/cmd.glob..func4(0x104985960?, {0x14000214ba0?, 0x3?, 0x3?})
	/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/transcribe.go:259 +0xafc
github.com/spf13/cobra.(*Command).execute(0x104985960, {0x14000214b70, 0x3, 0x3})
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:876 +0x4b8
github.com/spf13/cobra.(*Command).ExecuteC(0x1049856e0)
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:990 +0x354
github.com/spf13/cobra.(*Command).Execute(...)
	/home/runner/work/assemblyai-cli/assemblyai-cli/vendor/github.com/spf13/cobra/command.go:918
github.com/AssemblyAI/assemblyai-cli/cmd.Execute()
	/home/runner/work/assemblyai-cli/assemblyai-cli/cmd/root.go:42 +0xa8
main.main()
	/home/runner/work/assemblyai-cli/assemblyai-cli/main.go:11 +0x1c

Not sure why it happened, I tried with different brazilian videos, and same thing. Maybe it is something about the language.
When trying with ensligh videos, everything works fine.