API para X / Twitter: Transcrição do tweet
Extraia a transcrição da fala de um tweet com vídeo do Twitter/X pela URL.
US$ 1,20/mil req., pague por requisição, sem assinatura.
Post
buildwithjonah
@buildwithjonah
Transcript
Alright so I spent the last six months building a data pipeline that normalizes about forty different scraper APIs into one schema. I want to walk through what broke and what I would do differently if I started over today. The first lesson was that every provider's error format is completely different. Some return a 200 with an error field buried three levels deep, some return a real HTTP error code, and some just silently return an empty array. We ended up writing a translation layer just so our retry logic actually knew whether a failure meant retry later or never retry at all. The second lesson was around pagination. Cursors are never portable across providers, so we had to build our own cursor abstraction just to keep the client code simple and predictable.
{"transcript": "Alright so I spent the last six months building a data pipeline that normalizes about forty different scraper APIs into one schema. I want to walk through what broke and what I would do differently if I started over today. The first lesson was that every provider's error format is completely different. Some return a 200 with an error field buried three levels deep, some return a real HTTP error code, and some just silently return an empty array. We ended up writing a translation layer just so our retry logic actually knew whether a failure meant retry later or never retry at all. The second lesson was around pagination. Cursors are never portable across providers, so we had to build our own cursor abstraction just to keep the client code simple and predictable."}
Preços e provedores
Preços do endpoint Transcrição do tweet da API para X / Twitter: uma API, 2 provedores intercambiáveis
O provedor mais barato atende primeiro. Se ele falhar, o próximo assume na mesma chamada.
Como chamar
Obtenha Transcrição do tweet do X / Twitter em uma só requisição
import os, requests res = requests.post( "https://api.getanyapi.com/v1/run/twitter.tweet_transcript", headers={"Authorization": f"Bearer {os.environ['ANYAPI_KEY']}"}, json={ "url": "https://x.com/TheoVon/status/1916982720317821050" }, ) print(res.json())
Este exemplo mostra 3 de 4 campos. Ver todos os campos
{
"found": true,
"data": {
"transcript": "Alright so I spent the last six months building a data pipeline that normalizes about forty different scraper APIs into one schema. I want to walk through what broke and what I would do differently if I started over today. The first lesson was that every provider's error format is completely different. Some return a 200 with an error field buried three levels deep, some return a real HTTP error code, and some just silently return an empty array. We ended up writing a translation layer just so our retry logic actually knew whether a failure meant retry later or never retry at all. The second lesson was around pagination. Cursors are never portable across providers, so we had to build our own cursor abstraction just to keep the client code simple and predictable."
}
}Referência completa de parâmetros e resposta - todos os campos, tipos e exemplos deste endpoint.
Referência
Uma requisição, um formato de resposta.
Cada campo que você envia e cada campo que você recebe, cada um com um exemplo real.
Última verificação em 2026-11-14 · disponibilidade e latência medidas em 30d
Corpo da requisição
JSON, enviado via POST para este endpoint.
- urlstring
https://x.com/buildwithjonah/status/1842093771280384512
URL do tweet com o vídeo a transcrever (ex.: https://x.com/TheoVon/status/1916982720317821050).
Resposta
JSON, um valor de exemplo por campo.
- transcriptstring
Alright so I spent the last six months building a data pipeline that normalizes about forty different scraper APIs into one schema. I want to walk through what broke and what I would do differently if I started over today. The first lesson was that every provider's error format is completely different. Some return a 200 with an error field buried three levels deep, some return a real HTTP error code, and some just silently return an empty array. We ended up writing a translation layer just so our retry logic actually knew whether a failure meant retry later or never retry at all. The second lesson was around pagination. Cursors are never portable across providers, so we had to build our own cursor abstraction just to keep the client code simple and predictable.
- foundboolean
true
- reason"not_found" | "too_long"
Present only when `found` is false, and says why there is no result. `not_found`: the source states the target does not exist, or returned nothing for it. `too_long`: the video is longer than the source can transcribe, so asking again returns the same answer. A `found: false` answer is a successful call, not an error, and `costUsd` is what it actually cost.
Dúvidas frequentes
Sobre o endpoint Transcrição do tweet da API para X / Twitter
O endpoint Transcrição do tweet da AnyAPI para X / Twitter retorna dados de X / Twitter em JSON normalizado com uma chamada POST para /v1/run/twitter.tweet_transcript. Extraia a transcrição da fala de um tweet com vídeo do Twitter/X pela URL. A AnyAPI distribui cada requisição entre 2 fontes, com failover automático quando uma falha. Custa a partir de US$ 1,20 por mil requisições, em dólares, sem assinatura e sem mínimo mensal. Nos últimos 30 dias, 99,9% das chamadas ao endpoint Transcrição do tweet da API para X / Twitter feitas pela AnyAPI tiveram sucesso, com tempo de resposta mediano de 4,4 segundos, em 900 chamadas medidas.