C# Calling OpenAI to process audio file doesn't work. With HttpClient or OpenAI NuGet

Question

Getting really frustrated with this. Really wish OpenAI would have some sort of technical help, but whatever.

I am trying to test to see if we can upload a WAV file and have it translate to text. Their API Documentation indicates they can with the following endpoint: POST https://api.openai.com/v1/audio/transcriptions

Of course there are no code samples using C#, but trying to modify the node one to work with either HttpClient or OpenAI NuGet package, it keeps throwing errors saying bad request. I even went so far to see if ChatGPT and render the code, which that too fails lol lots of help that is.

For the most part, I've tried this variation:

static async Task<string> ConvertAudioToText(string filePath, string apiKey)
{
    using var httpClient = new HttpClient();
    httpClient.DefaultRequestHeaders.Add("Authorization", $"Bearer {apiKey}");
    httpClient.DefaultRequestHeaders.Accept.Add(new MediaTypeWithQualityHeaderValue("application/json"));

    var audioBytes = await File.ReadAllBytesAsync(filePath);

    using var content = new MultipartFormDataContent();
    using var audioContent = new ByteArrayContent(audioBytes);
    audioContent.Headers.ContentType = new MediaTypeHeaderValue("audio/wav");
    content.Add(audioContent, "model", "whisper-1");

    var response = await httpClient.PostAsync("https://api.openai.com/v1/engines/whisper-1/transcriptions", content);
    var responseContent = await response.Content.ReadAsStringAsync();

    return responseContent;
}

Which keeps returning this error message:

"error": {
    "message": "you must provide a model parameter",
    "type": "invalid_request_error",
    "param": null,
    "code": null
}

Any help trying to figure this out, would be great. Any idea why this gives me an error, I provided the model but it keeps asking for it. How do I provide a file and model types?

Thanks!

Answer 1

This Should help you. This is an example of call to get the transcript as SRT file, from an audio (or video) file.

HttpClient client = new HttpClient();
string filePath = "path_to_your_file";

client.DefaultRequestHeaders.Authorization = new AuthenticationHeaderValue("Bearer", "YOUR_API_KEY");

using (var memoryStream = new MemoryStream())
{
    using (var fs = File.OpenRead(filePath))
    {
        fs.CopyTo(memoryStream);
        var content = new MultipartFormDataContent
        {
            { new StringContent("whisper-1"), "model" },
            { new StringContent("srt"), "response_format" },
            { new ByteArrayContent(memoryStream.ToArray()), "file", filePath }
        };
        var response = await client.PostAsync("https://api.openai.com/v1/audio/transcriptions", content);
        var transcript = await response.Content.ReadAsStringAsync();
    }
}