Documentation
Table of contents
Before you start
Unstructured data extraction
Content summarization
Key insights extraction
Spam detection
Authentication
Authenticating your requests with Nobinge's API couldn't be easier. Please follow these steps:
- Subscribe to a plan.
- Create an API token.
-
Use it as a bearer token in the
Authorization
header of your requests.
Request
POST https://nobinge.ai/api/v1/the-endpoint-you-choose Authorization: Bearer YOUR_API_TOKEN { "Foo": "Bar" }
Important notes
- A lot of computing is happening each time you call Nobinge's API. Expect multiple tens of seconds per response depending on the amount of information that needs to be processed.
- Our homemade web crawler is good. But it doesn't solve captchas (yet) and some websites (like Amazon for instance) have really good bot protections we can't bypass (for now at least). If you are paying for a scraping service, we recommend using it alongside Nobinge's API.
Extract data from a webpage
The endpoint will return a JSON structure containing the data you need. You have to specify the columns first.
For instance, if you want to extract data from Rick Astley's Wikipedia page, you can ask for columns such as "Name", "Description", "Birth Date", "Children Count", "Years Active", etc.
Please note that we leverage our own crawler for this endpoint. If you have for habit to use another one, please feel free to do so and call the "Extract data from text" endpoint.
Parameters
url
string
required
The URL of the webpage you want to extract data from.
columns
array
required
The columns you need.
as_text
boolean
optional
Sometimes, HTML content is too lengthy for the AI. While the limit is pushed back year after year, getting the content of a page as text may be necessary in the meantime.
false
by default.
Request
POST https://nobinge.ai/api/v1/extract-data-from-webpage Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "url": "https://en.wikipedia.org/wiki/Rick_Astley", "columns": [ "Name", "Description", "Birth Date", "Children Count", "Years Active" ] }
Response
{ "data": { "Name": "Rick Astley", "Description": "Richard Paul Astley (born 6 February 1966) is an English singer. He gained fame through his association with the production trio Stock Aitken Waterman, releasing the 1987 album 'Whenever You Need Somebody,' which sold 15 million copies worldwide. His debut single 'Never Gonna Give You Up' was a No. 1 hit in 25 countries, winning the 1988 Brit Award for Best British Single. His 1988 single 'Together Forever' became his second song to top the US Billboard Hot 100 and was one of his eight songs to reach the Top 10 on the UK Singles Chart. By the time of his brief retirement, Astley had sold approximately 40 million records worldwide. He became an internet phenomenon in 2007 when the music video for 'Never Gonna Give You Up' became integral to the Rickrolling meme, and his performance career was revitalized by the meme's popularity.", "Birth Date": "1966-02-06", "Children Count": 1, "Years Active": "1985\u20131993, 2000\u2013present" } }
Extract data from text
The endpoint will return a JSON structure containing the data you need. You have to specify the columns first.
For instance, if you want to extract data from the content of Rick Astley's Wikipedia page, you can ask for columns such as "Name", "Description", "Birth Date", "Children Count", "Years Active", etc.
Parameters
text
string
required
The content you want to extract data from.
columns
array
required
The columns you need.
Request
POST https://nobinge.ai/api/v1/extract-data-from-text Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "text": "Richard Paul Astley (born 6 February 1966) is an English singer. He gained fame through his association with the production trio Stock Aitken Waterman, releasing the 1987 album Whenever You Need Somebody…", "columns": ["Name", "Description", "Birth Date", "Children Count", "Years Active"] }
Response
{ "data": { "Name": "Rick Astley", "Description": "Richard Paul Astley (born 6 February 1966) is an English singer. He gained fame through his association with the production trio Stock Aitken Waterman, releasing the 1987 album 'Whenever You Need Somebody,' which sold 15 million copies worldwide. His debut single 'Never Gonna Give You Up' was a No. 1 hit in 25 countries, winning the 1988 Brit Award for Best British Single. His 1988 single 'Together Forever' became his second song to top the US Billboard Hot 100 and was one of his eight songs to reach the Top 10 on the UK Singles Chart. By the time of his brief retirement, Astley had sold approximately 40 million records worldwide. He became an internet phenomenon in 2007 when the music video for 'Never Gonna Give You Up' became integral to the Rickrolling meme, and his performance career was revitalized by the meme's popularity.", "Birth Date": "1966-02-06", "Children Count": 1, "Years Active": "1985\u20131993, 2000\u2013present" } }
Summarize a YouTube video
Summarize any YouTube video with transcripts enabled. The API will return the summary as well as some basic information about the video that you can use.
Parameters
url
string
required
The URL of the YouTube video you want to summarize. It can be in any format (www.youtube.com, youtu.be). the API will extract the video ID from it.
level
string
optional
The level of summarization you want. It can be concise, balanced, or extensive. Default is balanced.
preferred_language
string
optional
In rare cases, the default transcript language of a video is not the one you expect. Or you may want to summarize the video in another language.
Possible values: af, ak, sq, am, ar, hy, as, ay, az, bn, eu, be, bho, bs, bg, my, ca, ceb, zh-Hans, zh-Hant, co, hr, cs, da, dv, nl, en, eo, et, ee, fil, fi, fr, gl, lg, ka, de, el, gn, gu, ht, ha, haw, iw, hi, hmn, hu, is, ig, id, ga, it, ja, jv, kn, kk, km, rw, ko, kri, ku, ky, lo, la, lv, ln, lt, lb, mk, mg, ms, ml, mt, mi, mr, mn, ne, nso, no, ny, or, om, ps, fa, pl, pt, pa, qu, ro, ru, sm, sa, gd, sr, sn, sd, si, sk, sl, so, st, es, su, sw, sv, tg, ta, tt, te, th, ti, ts, tr, tk, uk, ur, ug, uz, vi, cy, fy, xh, yi, yo, zu
Request
POST https://nobinge.ai/api/v1/summarize-youtube-video Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "url": "https://www.youtube.com/watch?v=some-id" }
Response
{ "data": { "video_id": "some-youtube-video-id", "channel_id": "some-channel-id", "channel_name": "Some Channel Name", "title": "Some YouTube Video Title", "summary": "Do proident esse tempor incididunt aute adipisicing ut enim. Veniam enim excepteur excepteur elit consequat aliquip sit. Sit magna laborum laborum reprehenderit consectetur eiusmod enim id minim voluptate. In laboris occaecat pariatur esse aliqua velit duis dolore minim proident. Sit enim eiusmod culpa fugiat reprehenderit consequat duis ea incididunt. Incididunt dolor magna aliquip incididunt." } }
Get key insights from text
Get a given amount of key insights from arbitrary content.
Parameters
text
string
required
The content you wish to get keyinsights from.
amount
integer
optional
The amount of key insights you need. Default is 3.
Request
POST https://nobinge.ai/api/v1/get-key-insights-from-text Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "text": "Richard Paul Astley (born 6 February 1966) is an English singer. He gained fame through his association with the production trio Stock Aitken Waterman, releasing the 1987 album Whenever You Need Somebody…", "amount": 5 }
Response
{ "data": { "key_insights": [ "Rick Astley was born on February 6, 1966.", "His debut single 'Never Gonna Give You Up' topped charts.", "Astley sold approximately 40 million records worldwide.", "He became famous through the Rickrolling internet meme.", "Astley's latest album 'Are We There Yet?' released in 2023." ] } }
Summarize a webpage
Summarize any webpage. The API will return the summary as well as some basic information about the page that you can use. An API credit will be consumed for each successful request (any HTTP status code lower than 500).
Please note that we leverage our own crawler for this endpoint. If you have for habit to use another one, please feel free to do so and call the "Summarize content" endpoint.
Parameters
url
string
required
The URL of the webpage you want to summarize.
level
string
optional
The level of summarization you want. It can be concise, balanced, or extensive. Default is balanced.
Request
POST https://nobinge.ai/api/v1/summarize-webpage Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "url": "https://example.com/foo" }
Response
{ "data": { "url": "https://example.com/foo", "title": "Some Title", "author": "Some Author", "summary": "Esse sit aute commodo et enim tempor ullamco est officia fugiat. Aliqua ullamco consequat sint commodo ex nulla fugiat est in sint exercitation duis consectetur. Enim aliquip minim ipsum incididunt exercitation. Culpa enim duis voluptate Lorem ipsum est Lorem ullamco exercitation sunt esse. Reprehenderit laborum nisi sit occaecat sint aliquip aute. Sit ipsum occaecat eiusmod consectetur laboris culpa pariatur." } }
Summarize content
Summarize any textual content you want. The API will simply return the summary.
Parameters
text
string
required
The content you wish to summarize.
level
string
optional
The level of summarization you want. It can be concise, balanced, or extensive. Default is balanced.
Request
POST https://nobinge.ai/api/v1/extract-data-from-webpage Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "text": "Officia velit ipsum ad nisi eu ut id eu voluptate. Elit culpa do sunt Lorem nostrud. Pariatur aliqua et exercitation laboris ut qui ex aute dolor exercitation quis nisi in. Esse adipisicing voluptate incididunt pariatur Lorem esse. Pariatur aliquip mollit ut qui dolor id dolor dolor duis consequat consectetur officia magna. Fugiat ad aliqua amet reprehenderit. Amet laborum et sint aliqua fugiat cillum esse ad velit. Nulla laborum ad velit culpa ipsum.", "level": "short" }
Response
{
"data": {
"summary": "Labore Lorem velit tempor culpa eu aliquip laboris veniam est nisi deserunt proident non. Excepteur nisi cillum mollit labore officia adipisicing ad voluptate labore proident culpa nostrud."
}
}
Evaluate webpage for spam
Evaluate a webpage for spam. The endpoint will return a score and an explanation. 0 is legitimate content, a value between 0 and 1 will be up to your interpretation, and 1 is spam.
Parameters
url
string
required
The URL of the webpage you want to evaluate.
Request
POST https://nobinge.ai/api/v1/is-webpage-spam Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "url": "https://adult-site-example.com", }
Response
{ "data": { "spam_score": 1, "explanation": "The content primarily promotes a premium subscription service for an adult site, offering free access for a limited time. It contains multiple calls to action encouraging users to sign up for premium access, which is commercial in nature. The repeated emphasis on free access and exclusive content, along with the promotional links, indicates a strong commercial intent without providing substantial informational value. Therefore, it is flagged as spam." } }
Evaluate content for spam
Evaluate any content for spam. The endpoint will return a score and an explanation. 0 is legitimate content, a value between 0 and 1 will be up to your interpretation, and 1 is spam.
Parameters
text
string
required
The content you want to evaluate.
Request
POST https://nobinge.ai/api/v1/is-content-spam Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "text": "Get rich quick! http://example-scam.xyz", }
Response
{ "data": { "spam_score": 1, "explanation": "The content promotes a 'get rich quick' scheme and includes a link to a suspicious website that appears to be a scam. This type of content is typically associated with spam and lacks any informative or valuable context." } }
Get the transcript of a YouTube video
Fetch the transcript of YouTube videos that have it enabled. This endpoint is unlimited for paying customers.
Parameters
url
string
required
The URL of the YouTube video you want to summarize. It can be in any format (www.youtube.com, youtu.be). the API will extract the video ID from it.
preferred_language
string
optional
In rare cases, the default transcript language of a video is not the one you expect. Or you may want to get the transcript in another language.
Possible values: af, ak, sq, am, ar, hy, as, ay, az, bn, eu, be, bho, bs, bg, my, ca, ceb, zh-Hans, zh-Hant, co, hr, cs, da, dv, nl, en, eo, et, ee, fil, fi, fr, gl, lg, ka, de, el, gn, gu, ht, ha, haw, iw, hi, hmn, hu, is, ig, id, ga, it, ja, jv, kn, kk, km, rw, ko, kri, ku, ky, lo, la, lv, ln, lt, lb, mk, mg, ms, ml, mt, mi, mr, mn, ne, nso, no, ny, or, om, ps, fa, pl, pt, pa, qu, ro, ru, sm, sa, gd, sr, sn, sd, si, sk, sl, so, st, es, su, sw, sv, tg, ta, tt, te, th, ti, ts, tr, tk, uk, ur, ug, uz, vi, cy, fy, xh, yi, yo, zu
Request
POST https://nobinge.ai/api/v1/get-youtube-video-transcript Authorization: Bearer YOUR_API_TOKEN Accept: application/json { "url": "https://www.youtube.com/watch?v=some-id" }
Response
{ "data": [ "url": "https://www.youtube.com/watch?v=some-youtube-video-id", "transcript": "Aute incididunt nostrud aute amet ullamco ullamco qui nulla mollit minim nisi proident fugiat elit…" ] }