gl_speech request almost always times out, no proper error message about googlelanguager HOT 18 CLOSED

My Session info:

sessionInfo()
R version 3.6.3 (2020-02-29)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Catalina 10.15.5

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] googleLanguageR_0.3.0

loaded via a namespace (and not attached):
[1] magrittr_1.5 R6_2.4.1 rlang_0.4.6 fansi_0.4.1 httr_1.4.1 tools_3.6.3 cli_2.0.2
[8] googleAuthR_1.3.0 askpass_1.1 ellipsis_0.3.1 openssl_1.4.2 assertthat_0.2.1 digest_0.6.25 tibble_3.0.1
[15] gargle_0.5.0 lifecycle_0.2.0 crayon_1.3.4 purrr_0.3.4 base64enc_0.1-3 vctrs_0.3.1 fs_1.4.2
[22] curl_4.3 memoise_1.1.0 glue_1.4.1 compiler_3.6.3 pillar_1.4.4 jsonlite_1.7.0 pkgconfig_2.0.3

from googlelanguager.

Looks like something underlying the API or library, are you behind a proxy perhaps? Perhaps check that the authentication process is correct by running some of the other functions such as gl_translate() examples. If you run the code with options(googleAuthR.verbose=2) we will get a bit more logging info to try and deduce what is happening.

from googlelanguager.

Whoah. what a fast answer.

Nop I'm not behind a proxy. Will Check the more verbosed output first.

from googlelanguager.

Ok with the verbosed output I'm getting:

gl_speech(test_audio)$transcript
ℹ 2020-07-02 10:48:05 > Token exists.
ℹ 2020-07-02 10:48:05 > Request: https://speech.googleapis.com/v1/speech:recognize/
ℹ 2020-07-02 10:48:05 > Body JSON parsed to: {"config":{"encoding":"LINEAR16","sampleRateHertz":16000,"languageCode":"en-US","maxAlternatives":1,"profanityFilter":false,"speechContexts":{},"enableWordTimeOffsets":true},"audio":{"content": [base code audio i think]

and after that nothing happens.

Will check the other apis but have to enable them first.

from googlelanguager.

Ok I tried it out with gl_translate and that works flawlessly:

gl_translate(text, target = "da")$translatedText
2020-07-02 10:55:58 -- Translating text: 114 characters -
ℹ 2020-07-02 10:55:58 > Token exists.
ℹ 2020-07-02 10:55:58 > Request: https://translation.googleapis.com/language/translate/v2/
[1] "at administrere medicin til dyr er ofte en meget vanskelig sag, og alligevel er det undertiden nødvendigt at gøre det"

translate British into Danish

It seems to have something to do with the speech to text api.

from googlelanguager.

Hmm ok thanks, it looks ok so nothing obvious, I'll investigate, perhaps test the example audio just in case.

One thing you may try also is use another PC or rstudio.cloud to see if you get the same issue, to see if we can see if its something to do with your setup.

from googlelanguager.

Good Idea will change to the setup of my colleague.

FYI:
I tried it wit R 4.0.2 as well as 3.6.3. R-studio version is 1.3.959

from googlelanguager.

Could you also report your sessionInfo() ? Mine is below, which I've confirmed works with the code in your issue:

> sessionInfo()
R version 3.6.3 (2020-02-29)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Catalina 10.15.5

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] googleLanguageR_0.3.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.4.6           rstudioapi_0.11        googleAnalyticsR_0.8.0
 [4] magrittr_1.5           tidyselect_1.1.0       R6_2.4.1              
 [7] rlang_0.4.6            fansi_0.4.1            httr_1.4.1            
[10] dplyr_1.0.0            tools_3.6.3            packrat_0.5.0         
[13] utf8_1.1.4             cli_2.0.2              withr_2.1.2           
[16] googleAuthR_1.3.0      askpass_1.1            cranlogs_2.1.1        
[19] ellipsis_0.3.0         openssl_1.4.1          assertthat_0.2.1      
[22] digest_0.6.25          tibble_3.0.0           gargle_0.4.0          
[25] lifecycle_0.2.0        crayon_1.3.4           tidyr_1.0.2           
[28] purrr_0.3.3            base64enc_0.1-3        vctrs_0.3.1           
[31] fs_1.4.1               curl_4.3               memoise_1.1.0         
[34] glue_1.4.1             compiler_3.6.3         pillar_1.4.3          
[37] generics_0.0.2         jsonlite_1.6.1         pkgconfig_2.0.3

from googlelanguager.

Ok I tried it out on my colleagues machine, same result :(

sessionInfo()
R version 3.5.1 (2018-07-02)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS 10.15.4

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] googleLanguageR_0.3.0

loaded via a namespace (and not attached):
[1] Rcpp_1.0.3 fansi_0.4.1 digest_0.6.16 crayon_1.3.4 assertthat_0.2.0
[6] R6_2.2.2 jsonlite_1.6.1 magrittr_1.5 pillar_1.4.3 httr_1.4.1
[11] rlang_0.4.5 cli_2.0.2 curl_3.2 rstudioapi_0.11 fs_1.3.2
[16] googleAuthR_1.3.0 tools_3.5.1 glue_1.3.0 purrr_0.3.3 yaml_2.2.0
[21] compiler_3.5.1 askpass_1.1 pkgconfig_2.0.2 base64enc_0.1-3 gargle_0.5.0
[26] memoise_1.1.0 openssl_1.4.1 tibble_2.1.3

Still a time out error. I also tried every step, routing my traffic over an vpn (to make sure it is not a internet problem) but didn't change anything.

from googlelanguager.

Hmm really don't know where to start sorry, you have the same setup as me and the auth works for other services - I can only think its some kind of firewall or network issue if you are using the same connection as your colleague.

from googlelanguager.

Example audio file also works with out problems. Checked that now.

from googlelanguager.

Oh the example audio file works, but the audio file you are trying doesn't? In that case it must be the audio encoding of the file?

from googlelanguager.

Does your file work on the website demo? https://cloud.google.com/speech-to-text

from googlelanguager.

Nop, the example files work if played in quicktime.

In general I don't get the error every time. Every now and than (maybe in 5% of the cases) I also got the text back from the gl_speech call

from googlelanguager.

Ok so the example audio file works 5% of cases?

But the audio file you are trying to transcribe never works?

It says to me its network connection issues, it is uploading a big chunk of audio that perhaps your corporate firewall blocks (most of the time) - you could try uploading your audio to Google Cloud Storage instead and using its URI so it has a secure upload. Also your example audio needs proper encoding (of which there are millions) - if you try the audio in the website perhaps you can see if it accepts it at all. There is more on audio encoding here: https://cloud.google.com/speech-to-text/docs/encoding

from googlelanguager.

Yes, the example file works in around 5% of the cases and my own files never work.
I coded my own file as wav, but also tried out flac and mp3 before.

If I try both files (example and my own, both as wav) on the demo webpage the work without any problems (besides asking me if i'm a robot every time).

Actually I don't think the the firewall is a problem. I'm in Home Office at the moment and here I have a standard DSL connection and if I use the VPN, I'm connected with the DFN (German research network) which is by default extremely relaxed in there firewall setting.

I now have tried out the demo webpage with my own audio file and the webpage also fails some times and gives me an "unauthorized" or a "server errror" message.

from googlelanguager.

It really seems to be somehow due to the transmission. I have now uploaded the data to a bucket in the gcloud and now it works.
I couldn't get it to work directly from the client. In the meantime I would have thought that there was a problem with the encoding of the audio file in json, because my computer was 100% busy on one core while trying to execute the request.

from googlelanguager.

Ok cool, glad its working now at least. If you want to automate the uploads there is cloudyr/googleCloudStorageR which will give you the URI for use in gl_speech()

from googlelanguager.