1
votes

When I tried to upload large file (from observation is greater than 8MB, around 3-4mins), I got "request is aborted" error.

Anyone experiencing similar issue?

I am using C# and uses the HTTP REST Interface of Speech to Text using sessionless method.

This is the url I am hitting: https://stream.watsonplatform.net/speech-to-text/api/v1/recognize?timestamps=true≺ofanity_filter=false&max_alternatives=3&word_confidence=true&word_alternatives_threshold=0.7&model=en-US_NarrowbandModel&continuous=true∈activity_timeout=600

I tried both using "Transfer-Encoding: chunked" and without.

Please help. Thanks.

2
Hello friend, can you take a look at some of the sample code we provide? we currently do not have any sample code for C# but maybe you can get an inspiration for the python client (github.com/watson-developer-cloud/…) or the Java SDK (github.com/watson-developer-cloud/java-sdk):Daniel Bolanos
Also, as a first step, did you try with curl? see an example command here: ibm.com/watson/developercloud/doc/speech-to-text/tutorial.shtmlDaniel Bolanos

2 Answers

0
votes

Split on chunks and upload separately. Or use other services like speechmatics or even open soruce implementation, they work with very large files.

0
votes

you can push files/streams of up to 100MBs in a single request using Watson STT API. If you are using C# please take a look at the .NET SDK, that may be a good start for you: https://github.com/watson-developer-cloud/dotnet-standard-sdk. And, of course, nothing like getting started with a curl command to build confidence, see details here: https://www.ibm.com/watson/developercloud/doc/speech-to-text/getting-started.html