OpenAI Node lib error on Audio Transcription

API
1

Im trying to record audio from user and send to whisper to transcribe, but the openai node lib is returning Axios error with 400 bad request.

response: {
    status: 400,
    statusText: 'Bad Request',
    headers: {
      date: 'Sat, 15 Apr 2023 16:08:50 GMT',
      'content-type': 'application/json',
      'content-length': '204',
      connection: 'close',
      'openai-organization': 'user-',
      'openai-processing-ms': '117',
      'openai-version': '2020-10-01',
      'strict-transport-security': 'max-age=15724800; includeSubDomains',
      'x-ratelimit-limit-requests': '50',
      'x-ratelimit-remaining-requests': '49',
      'x-ratelimit-reset-requests': '1.2s',
      'x-request-id': '36c8e2b87112c8e80dc44ee652344c25',
      'cf-cache-status': 'DYNAMIC',
      server: 'cloudflare',
      'cf-ray': '7b85730edf68a5d4-GRU',
      'alt-svc': 'h3=":443"; ma=86400, h3-29=":443"; ma=86400'
    },
    config: {
      transitional: [Object],
      adapter: [Function: httpAdapter],
      transformRequest: [Array],
      transformResponse: [Array],
      timeout: 0,
      xsrfCookieName: 'XSRF-TOKEN',
      xsrfHeaderName: 'X-XSRF-TOKEN',
      maxContentLength: -1,
      maxBodyLength: -1,
      validateStatus: [Function: validateStatus],
      headers: [Object],
      method: 'post',
      data: [FormData],
      url: 'https://api.openai.com/v1/audio/transcriptions'
    },
    request: <ref *1> ClientRequest {
      _events: [Object: null prototype],
      _eventsCount: 7,
      _maxListeners: undefined,
      outputData: [],
      outputSize: 0,
      writable: true,
      destroyed: false,
      _last: true,
      chunkedEncoding: true,
      shouldKeepAlive: false,
      maxRequestsOnConnectionReached: false,
      _defaultKeepAlive: true,
      useChunkedEncodingByDefault: true,
      sendDate: false,
      _removedConnection: false,
      _removedContLen: false,
      _removedTE: false,
      strictContentLength: false,
      _contentLength: null,
      _hasBody: true,
      _trailer: '',
      finished: true,
      _headerSent: true,
      _closed: false,
      socket: [TLSSocket],
      _header: 'POST /v1/audio/transcriptions HTTP/1.1\r\n' +
        'Accept: application/json, text/plain, */*\r\n' +
        'Content-Type: multipart/form-data; boundary=--------------------------5047590565482031\r\n' +
        'User-Agent: OpenAI/NodeJS/3.2.1\r\n' +
        'Authorization: Bearer sk-\r\n' +
        'Host: api.openai.com\r\n' +
        'Connection: close\r\n' +
        'Transfer-Encoding: chunked\r\n' +
        '\r\n',
      _keepAliveTimeout: 0,
      _onPendingData: [Function: nop],

This is how im requesting:

import FormData from 'form-data';
import { withFileUpload } from 'next-multiparty';
import { createReadStream, read } from 'fs';

const fs = require('fs');

export const config = {
  api: {
    bodyParser: false,
  },
};

const { Configuration, OpenAIApi } = require("openai");
const configuration = new Configuration({
  apiKey: process.env.OPENAI_API_KEY,
});
const openai = new OpenAIApi(configuration);

export default withFileUpload(async (req, res) => {

  const file:any= fs.createReadStream(req.file?.filepath);

  const response = await openai.createTranscription(file,"whisper-1").then((res) => console.log(res.data)).catch((err) => console.log(err));

  if (response) {
    return response.text;
  }else{
    return "Speech to text failed"
}

});

I tried looking on github, but this error keeps happening.

1 Like
2

I think that it is not problem with your code.
I tried to do that with curl as described here: OpenAI API
curl --request POST --url https://api.openai.com/v1/audio/transcriptions --header “Authorization: Bearer APIKEY” --header “Content-Type: multipart/form-data” --form file=@out000.mp3 --form model=whisper-1
and it also returns:

400 Bad Request

400 Bad Request

cloudflare
3

Any update on this? I am getting the same issue trying to call the transcriptions API from a node application. At first I was getting the issue of unsupported file format (I’m trying to send an mp3), now all I get is a 400 Bad Request. I have tried using openai library and axios to make the request.

4

I made it work by sending to my backend the audio in base64 and then converting it back into a mp3 file and sending the entire file to whisper

1 Like
5

Hey @thegen, can I please see your code? I tried to do the same with a m4a file but it didn’t work for me.

1 Like