We Keep Coding

Not compelete result from Azure speech to text API

I have a small wav sound file in which I want to get the text of it, so I used Azure speech to text API to test it.
first thing I convert the audio file as they recommended in their documentation to PCM - Mono -16K sample rate.
and I use this code in c# in the documentation example here to upload the file and get the result.
HttpWebRequest request = null;
request = (HttpWebRequest)HttpWebRequest.Create("https://speech.platform.bing.com/speech/recognition/interactive/cognitiveservices/v1?language=en-US&format=detailed");
request.SendChunked = true;
request.Accept = #"application/json;text/xml";
request.Method = "POST";
request.ProtocolVersion = HttpVersion.Version11;
request.ContentType = #"audio/wav; codec=audio/pcm; samplerate=16000";
request.Headers["Ocp-Apim-Subscription-Key"] = "my key";
// Send an audio file by 1024 byte chunks
using (FileStream fs = new FileStream("D:/b.wav", FileMode.Open, FileAccess.Read))
{
/*
* Open a request stream and write 1024 byte chunks in the stream one at a time.
*/
byte[] buffer = null;
int bytesRead = 0;
using (Stream requestStream = request.GetRequestStream())
{
/*
* Read 1024 raw bytes from the input audio file.
*/
buffer = new Byte[checked((uint)Math.Min(1024, (int)fs.Length))];
while ((bytesRead = fs.Read(buffer, 0, buffer.Length)) != 0)
{
requestStream.Write(buffer, 0, bytesRead);
}
// Flush
requestStream.Flush();
}
}
string responseString;
Console.WriteLine("Response:");
using (WebResponse response = request.GetResponse())
{
Console.WriteLine(((HttpWebResponse)response).StatusCode);
using (StreamReader sr = new StreamReader(response.GetResponseStream()))
{
responseString = sr.ReadToEnd();
}
Console.WriteLine(responseString);
Console.ReadLine();
}
also i tried using cUrl tool and also write it in java as i was thought that maybe it's problem with the programming language I use that i not upload the file correctly.
this the link of the sound file i want to convert it to text here.
so Now i need to help to figure it out if the problem comes from the format of the sound file? or from maybe code that i not upload it correctly? or it's from the API I mean to be not accurate enough?
i tried IBM speech to text and it got all the text with no problem.
iam using now the free trial of Azure speech to text API and I want to figure where the problem comes if anyone has experience with this to see if I will work with this API or not.
Update
I want to clear that iam not got any error i just got incomplete result to my sound file I upload, for example the sound file i upload he said at the end of the sound "What is up with that", the result i got from Azure is just the first sentence only which is "I say that like it's a bad thing.", also I upload another sound file which contains the "What is up with that" only check it here,and it just gives me an empty result like this.
{"RecognitionStatus":"NoMatch","Offset":17300000,"Duration":0}
so all that i want to know if this normal from the Speech to text API Azure or the problem with my code or from the sound file? this what i want to get an answer with it.
when i test another API on those files it worked like IBM for example.
Thanks in advance.

WebAPI to download PowerPoint file using memory stream

I'm having trouble downloading a file with content using a Powerpoint presentation library (Syncfusion). The docs only supply ASP examples, no Web API specifically.
I can save a file to the file system which has the context I add to the Powerpoint.
I can get the API to download a file to the users browser but this Powerpoint is empty.
Syncfusion has a function to save the Powerpoint to a memory stream so I guess my question is what is the correct way to save a file to the users browser with the content from the stream?
I'm using HTTPGet and hitting the link through the browser. Do I need to sent the context-type or anything like that?
Thanks for your help, I can provide what I have so far if that helps.
Kurtis
Edit:
[HttpGet, Route("")]
public HttpResponseMessage Get()
{
var presentation = Presentation.Create();
var firstSlide = presentation.Slides.Add(SlideLayoutType.Blank);
var textShape = firstSlide.AddTextBox(100, 75, 756, 200);
var paragraph = textShape.TextBody.AddParagraph();
paragraph.HorizontalAlignment = HorizontalAlignmentType.Center;
var textPart = paragraph.AddTextPart("Kurtis' Presentation");
textPart.Font.FontSize = 80;
var memoryStream = new MemoryStream();
presentation.Save(memoryStream);
var result = Request.CreateResponse(HttpStatusCode.OK);
result.Content = new StreamContent(memoryStream);
result.Content.Headers.ContentType = new MediaTypeHeaderValue("application/octet-stream");
result.Content.Headers.ContentDisposition = new ContentDispositionHeaderValue("attachment")
{
FileName = "export.pptx"
};
return result;
}
This is what I have, the library saves the presentation to the memory stream, you can change the parameter to a string which writes this to a file and there is an option to pass a filename, format and HttpResponse for MVC but I couldn't get this working with my API controller. I kept getting a network error but didn't know why.
Thanks again

I found my problem with the help of others in the comments.
The code above work but I needed to reset the stream position to zero to actually write the data.
memoryStream.Position = 0;
Thanks for those who commented. :-)

C# Uploading a memorystream object to Web API

I have an CSV file in memory that I want to upload to a Web API.
If I save the CSV file to disk and upload it, it gets accepted.
However, I want to avoid the extra work and also make the code cleaner by simply uploading the text I have as a MemoryStream Object (I think that's the correct format?).
The following code works for uploading the file:
string webServiceUrl = "XXX";
string filePath = #"C:\test.csv";
string cred = "YYY";
using (var client = new WebClient()){
client.Headers.Add("Authorization", "Basic " + cred);
byte[] rawResponse = client.UploadFile(webServiceUrl, "POST", filePath);
Console.WriteLine(System.Text.Encoding.ASCII.GetString(rawResponse));
}
How would I do if I had a string with all the contents and I want to upload it in the same way without having to save it down to a file?
WebClient.UploadData or WebClient.UploadString perhaps?
Thank you
EDIT:
I tried what you said but by using a local file (in case there was something wrong with the string), but I get the same error.
Here is what I suppose the code would be using your solution
string webServiceUrl = "XXX";
string file = #"C:\test.csv";
string cred = "YYY";
FileStream fs = new FileStream(file, FileMode.Open, FileAccess.Read);
BinaryReader r = new BinaryReader(fs);
byte[] postArray = r.ReadBytes((int)fs.Length);
using (var client = new WebClient())
{
client.Headers.Add("Authorization", "Basic " + cred);
using (var postStream = client.OpenWrite(webServiceUrl, "POST"))
{
postStream.Write(postArray, 0, postArray.Length);
}
}
Any thoughts?

Use OpenWrite() from the WebClient.
using (var postStream = client.OpenWrite(endpointUrl))
{
postStream.Write(memStreamContent, 0, memStream.Length);
}
As documentation mentioned:
The OpenWrite method returns a writable stream that is used to send data to a resource.
Update
Try to set the position of the MemoryStream to 0 before uploading.
memoryStream.Position = 0;
When you copy the file into the MemoryStream, the pointer is moved to the end of the stream, so when you then try to read it, you're getting a null byte instead of your stream data.
MSDN - CopyTo()
Copying begins at the current position in the current stream, and does not reset the position of the destination stream after the copy operation is complete.

I finally managed to solve it.
First I made a request using CURL that worked.
I analyzed the packet data and made an except copy of the packet.
I did a lot of changes, however, the final change was that using the different functions I found online it never closed the packet with a "Last-Boundary" while CURL did.
So by modifying the function, making sure it properly wrote a Last-Boundary it finally worked.
Also, another crucial thing was to set PreAuthenticate to true, the examples online didn't do that.
So, all in all:
1. Make sure that the packet is properly constructed.
2. Make sure you pre authenticate if you need to authenticate.
webrequest.PreAuthenticate = true;
webrequest.Headers[HttpRequestHeader.Authorization] = string.Format("Basic {0}", cred);
Don't forget to add SSL if using a https (which you probably do if you authenticate):
ServicePointManager.SecurityProtocol = SecurityProtocolType.Tls | SecurityProtocolType.Ssl3;
Hope this helps someone.
And thanks for the help earlier!

How to show download dialog before file is ready to download?

I'm getting a file from a database in byte [] format and want user to see download dialog before Linq will take it from the database. It's in C# and ASP.NET.
Now, it's like this:
User choose a file, click on it.
In code I get id of file clicked and using Linq I'm downloading.
Then I send the file by Response.OutputStream.Write(content, 0,
content.Length);
Before a file is downloaded from the database user won't see any
download dialog.
What can I do if I want users to see the download dialog before file is downloaded?
Code:
Getting file by id:
public static byte[] getFile(Guid id)
{
var linqFile = from file in MyDB.Files
where file.IdPliku.Equals(id)
select new
{
Content = file.Content
};
return linqFile.ToList().FirstOrDefault().Content.ToArray();
}
Saving file:
public void SaveFile(Guid fileID, string filename, string mimeTypes)
{
try
{
byte[] content = FileService.getFile(fileID);
Response.ClearContent();
Response.ClearHeaders();
Response.ContentType = mimeTypes;
Response.AppendHeader("Accept-Ranges", "bytes");
Response.AppendHeader("Content-Range", string.Format("0-{0}/{1}", content.Length, content.Length));
Response.AppendHeader("Content-Length", content.Length.ToString());
Response.AppendHeader("Content-Encoding", "utf-8");
Response.AppendHeader("Content-Type", Response.ContentType);
Response.AppendHeader("Content-Disposition", "attachment; filename= " + HttpUtility.UrlEncode(filename));
Response.OutputStream.Write(content, 0, content.Length);
//Response.BinaryWrite(content);
Response.Flush();
}
finally
{
Response.Close();
}
}
You are my hope.

your issue is here:
byte[] content = FileService.getFile(fileID);
because in this line you allocate the whole file in the web server's RAM and put everything in there, all content of the file from the database; what happens later does not matter anymore because you have already downloaded from db to web server in this line!!!
I am having such Deja-vu because I am sure I have given exactly the same comment on a very same question few weeks ago. Can't find it now, search for something like this here in SO.
In fact the solution is to stream directly to the output stream of the Response avoiding your byte[] array allocation above, to get this your data layer should of course support it and if it does not you could add a method for this. You want to use SQL Server filestream or something similar.

FileStream with querystring in filename?

I have a need to be able to open a file on disk but pass in parameters to that file via a querystring. It's a .SWF file, so I'm passing in the parameter necessary to get it to load correctly.
The code I'm using to do so is:
FileStream fs = new FileStream(#"C:\test\file.swf?key=value", FileMode.Open, FileAccess.Read);
I'm getting an error opening the file: "Invalid characters in path" because of the "?" in the filename. Is there any way to load a file from disk into a FileStream object using a querystring in the filename?

I think you can't do what you're trying to do. When you load a file from disk the querystring does not exist as a concept. It will only return the bytes contained in the SWF file.
The querystring matters at the execution level.

So I solved this problem by putting my two SWF files on a web server and using the following code. Not exactly production ready code, but it illustrates the concept.
private static FileStream GetFileStream()
{
string url = #"http://www.someurl.com/shell.swf?Filename=actualfile.swf";
HttpWebRequest request = (HttpWebRequest)HttpWebRequest.Create(url);
WebResponse response = request.GetResponse();
byte[] result = null;
int byteCount = Convert.ToInt32(response.ContentLength);
using (BinaryReader reader = new BinaryReader(response.GetResponseStream()))
result = reader.ReadBytes(byteCount);
return new FileStream(result);
}