Skip to content

Latest commit

 

History

History
73 lines (48 loc) · 2.59 KB

README.md

File metadata and controls

73 lines (48 loc) · 2.59 KB
Raw

Speech Recognition QuartzNet Python* Demo

This demo demonstrates Automatic Speech Recognition (ASR) with pretrained QuartzNet model.

How It Works

After computing audio features, running a neural network to get character probabilities, and CTC greedy decoding, the demo prints the decoded text.

Preparing to Run

The list of models supported by the demo is in <omz_dir>/demos/speech_recognition_quartznet_demo/python/models.lst file. This file can be used as a parameter for Model Downloader and Converter to download and, if necessary, convert models to OpenVINO Inference Engine format (*.xml + *.bin).

An example of using the Model Downloader:

omz_downloader --list models.lst

An example of using the Model Converter:

omz_converter --list models.lst

Supported Models

  • quartznet-15x5-en

NOTE: Refer to the tables Intel's Pre-Trained Models Device Support and Public Pre-Trained Models Device Support for the details on models inference support at different devices.

Running Demo

Run the application with -h option to see help message.

usage: speech_recognition_quartznet_demo.py [-h] -m MODEL -i INPUT [-d DEVICE]

optional arguments:
  -h, --help            Show this help message and exit.
  -m MODEL, --model MODEL
                        Required. Path to an .xml file with a trained model.
  -i INPUT, --input INPUT
                        Required. Path to an audio file in WAV PCM 16 kHz mono format
  -d DEVICE, --device DEVICE
                        Optional. Specify the target device to infer on, for
                        example: CPU, GPU, HDDL, MYRIAD or HETERO. The
                        demo will look for a suitable IE plugin for this
                        device. Default value is CPU.

The typical command line is:

python3 speech_recognition_quartznet_demo.py -m quartznet-15x5-en.xml -i audio.wav

NOTE: Only 16-bit, 16 kHz, mono-channel WAVE audio files are supported.

An example audio file can be taken from <openvino_dir>/deployment_tools/demo/how_are_you_doing.wav.

Demo Output

The application prints the decoded text for the audio file. The demo reports

  • Latency: total processing time required to process input data (from reading the data to displaying the results).

See Also