Httprequest to Deepgram - bug?

[thomaa]

Running httprequest node to deepgram doesn’t give the expected results.
Running the same httprequest on postman works!

It seems like I only receive parts of the result?

This is the result my workflow gives:

[
  {

(((SNIP)))

    "results": {
      "channels": [
        {
          "alternatives": [
            {
              "transcript": "",
              "confidence": 0,
              "words": [],
              "paragraphs": {
                "transcript": "\n",
                "paragraphs": []
              }
            }
          ]
        }
      ]
    }
  }
]

The transcript is empty.
This is what it looks like on postman:

(((SNIP)))
"results":{"channels":[{"alternatives":[{"transcript":"Men jeg kaller dere ikke lenger tjener dere, for tjener dere vet ikke hva Herren hans gjør. Jeg kaller dere venner. For jeg har gjort kjent for dere alt jeg har hørt av min far. Det står noe veldig interessant rett etterpå, som Jesu sier I hver 16 sted, og så står det her: Dere har ikke utvalgt meg, men jeg har utvalgt dere og satt dere til å gå ut og bære frukt, en frukt som varer. Og så sier Jesus","confidence":0.9873047,"words":[{"word":"men","start":0.16,"end":0.65999997,"confidence":0.71240234,"punctuated_word":"Men"},{"word":"jeg","start":1.5999999,"end":1.68,"confidence":0.9921875,"punctuated_word":"jeg"},{"word":"kaller","start":1.68,"end":1.92,"confidence":0.9667969,"punctuated_word":"kaller"},{"word":"dere","start":1.92,"end":2.08,"confidence":0.92333984,"punctuated_word":"dere"},{"word":"ikke","start":2.08,"end":2.32,"confidence":0.89990234,"punctuated_word":"ikke"},{"word":"lenger","start":2.32,"end":2.56,"confidence":0.9790039,"punctuated_word":"lenger"},{"word":"tjener","start":2.56,"end":2.72,"confidence":0.9838867,"punctuated_word":"tjener"},{"word":"dere","start":2.72,"end":2.96,"confidence":0.8027344,"punctuated_word":"dere,"},{"word":"for","start":2.96,"end":3.1999998,"confidence":0.9472656,"punctuated_word":"for"},{"word":"tjener","start":3.1999998,"end":3.36,"confidence":0.9902344,"punctuated_word":"tjener"},{"word":"dere","start":3.36,"end":3.6,"confidence":0.98095703,"punctuated_word":"dere"},{"word":"vet","start":3.6,"end":3.84,"confidence":0.9970703,"punctuated_word":"vet"},{"word":"ikke","start":3.84,"end":4.0,"confidence":0.93115234,"punctuated_word":"ikke"},{"word":"hva","start":4.0,"end":4.16,"confidence":0.99902344,"punctuated_word":"hva"}

(((SNIP)))
}]}}]}]}}

I have tried to call deepgram sync and async, but with same result.
This is the workflow:

Information on your n8n setup

• n8n version: 1.84.1
• Database (default: SQLite): -
• n8n EXECUTIONS_PROCESS setting (default: own, main): -
• Running n8n via (Docker, npm, n8n cloud, desktop app): cloud
• Operating system: -

[solomon]

Can you try copying the cURL from Postman and pasting it in n8n?
It looks like the requests are different. We just need to find where, exactly.

In Postman you can click this icon to copy the cURL

And then you paste the cURL in the HTTP Request node to copy the full settings from Postman to n8n:

.
If my reply answers your question, please remember to mark it as a solution.