https://github.com/scootpl/go-tensorflow-audio-example

An example of using a neural network model (LSTM) with the Tensorflow Go API.
https://github.com/scootpl/go-tensorflow-audio-example

audio audio-processing example golang guitar lstm-neural-networks machine-learning neural-network tensorflow tensorflow-go wav

Last synced: 10 months ago
JSON representation

An example of using a neural network model (LSTM) with the Tensorflow Go API.

• Host: GitHub
• URL: https://github.com/scootpl/go-tensorflow-audio-example
• Owner: scootpl
• License: mit
• Created: 2022-12-01T21:37:41.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2022-12-28T21:58:22.000Z (about 3 years ago)
• Last Synced: 2025-03-05T14:02:22.157Z (11 months ago)
• Topics: audio, audio-processing, example, golang, guitar, lstm-neural-networks, machine-learning, neural-network, tensorflow, tensorflow-go, wav
• Language: Go
• Homepage:
• Size: 1.15 MB
• Stars: 1
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

          
# Tensorflow Go API Audio Example

An example of using a neural network model (LSTM) with the Tensorflow Go API.

The model converts a clean guitar signal into an overdriven one, sonically similar to the Tube Screamer TS9

## Installation

Read documentation for installing the Go bindings for TensorFlow.

https://github.com/tensorflow/build/blob/master/golang_install_guide/README.md

## Input file

Input file can be in WAV format: 8, 16 or 24 bit (integer)

## Output file

Same as input

## Model

- model saved as: "saved_model.pb"

- input: 150 samples (float32)

- batch: multiple of 150 (default: 600)

## CLI Parameters

```

Usage of ./main:

  -i string

        input file name, "-" for Stdin

  -m string

        model directory

  -o string

        output file name

```

Model based on: https://github.com/GuitarML/GuitarLSTM