# Google Cloud Vision API - No value returned

**URL:** <https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076>\
**Category:** Ask for Help\
**Created:** [July 22, 2024, 3:52pm UTC](https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076 "2024-07-22T15:52:46Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![tom\_aera](https://avatars.discourse-cdn.com/v4/letter/t/65b543/32.png) [@tom\_aera](https://community.glideapps.com/u/tom_aera)\
**Post date:** [July 22, 2024, 3:52pm UTC](https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076/1 "2024-07-22T15:52:46Z")

</div>

Hello,

I want to use Cloud Vision to extract text from PDFs.  
However, I only get the error ‘No value returned from OCR Engine’.

![image](https://us1.discourse-cdn.com/flex002/uploads/glideapps/original/3X/1/4/1484f3122fac24211b839a7bc1cd59e97b8a1e23.png)

I tried different files. Is it because the input is a URL?

Thanks in advance!

Tom

---

<div class="post-metadata">

**Author:** ![ThinhDinh](https://sea2.discourse-cdn.com/flex002/user_avatar/community.glideapps.com/thinhdinh/32/49_2.png) [@ThinhDinh](https://community.glideapps.com/u/ThinhDinh)\
**Post date:** [July 22, 2024, 11:10pm UTC](https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076/2 "2024-07-22T23:10:38Z")

</div>

Does it have to be Cloud Vision? Glide has this “Document to Text” column that may fit what you need to do.

![image](https://us1.discourse-cdn.com/flex002/uploads/glideapps/original/3X/b/4/b48e916507fc571f7fb99620df85fc3ee58ec156.png)

---

<div class="post-metadata">

**Author:** ![tom\_aera](https://avatars.discourse-cdn.com/v4/letter/t/65b543/32.png) [@tom\_aera](https://community.glideapps.com/u/tom_aera)\
**Post date:** [July 23, 2024, 10:00am UTC](https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076/3 "2024-07-23T10:00:03Z")

</div>

Hi Thinh, thanks for the response!

That’s a great idea, yeah. But the AI is supposed to extract specific values from a pretty packed PDF. So I don’t know if extracting the text from it first will do a good job (since the values might be all over the place and taken out of context).

So I thought using an AI that analyses the whole PDF / ‘image’ at once would be better. Maybe the Image to Text module?

---

<div class="post-metadata">

**Author:** ![ThinhDinh](https://sea2.discourse-cdn.com/flex002/user_avatar/community.glideapps.com/thinhdinh/32/49_2.png) [@ThinhDinh](https://community.glideapps.com/u/ThinhDinh)\
**Post date:** [July 23, 2024, 11:13am UTC](https://community.glideapps.com/t/google-cloud-vision-api-no-value-returned/75076/4 "2024-07-23T11:13:03Z")

</div>

As far as I aware, it still has to extract the text first from your document/image, I don’t know how much “positioning” does it need to generate a correct answer. I trialled with something like that in the past, and it works with a bit of prompting (Document/Image to Text =\> Generate text with first step’s value being the input).