Speech to text model train & development

Job ID: 35201126

Budget: $400 – $500 USD

We want to develop a speech-to-text model for real-time stream audio to text recognition.

Requirements List

• Runs on linux server.

• The input is a stream audio, the output is json format recognition result, the json contains text result, words and words time stamp.

• English only

• Run multiple decoders, can process multiple audio streams on multi-core CPU.

• Performance, we have detailed performance requirements in our document, such as memory usage and process speed.

The above is all points and we have a detailed document for this task, we should implement it as per document.