Google’s annual developer Meet, d Google The I / O 2022 event recently ended. At this year’s event, the tech giant demonstrated multiple advances in new AI technology and security. Google’s Tone transfer Tool, one of these advances that was released in 2020 and reached the stage during the pre-show of the Google I / O 2022 event. This tool was created by Google Magenta The team and it allows users to turn ordinary instrumental sounds or even random sounds into something different.
What is Google Tone Transfer?
The Tone Transfer Tool was developed by Google’s Magenta team and was introduced by Google in 2020. This tool allows artists and creators to input different words and get something different For example, users can feed the instrument with sounds such as – people singing, birds chirping, or playing instruments and can convert them into the sound of musical instruments like flute or violin.
How does it work?
The Tone Transfer Tool project relies on a machine learning process called Differential digital signal processing Or DDSP. While testing the tool, the team introduced a violin with a 10-minute recording to process the model and run it through its system. Subsequently, the team began monitoring the progress of the program and its ability to recreate sound. The developers noticed that after an hour of training, the model was able to recreate the sound like a playing synth. Moreover, after 10 hours of training, the instrument was able to replicate the sound of a violin almost individually. The model can control the various audio signatures provided by the instrument to help it successfully reproduce the sound. This project shows how far machine learning has progressed as AI has acquired the ability to develop and reproduce musical instruments on its own.
How to create sound in tone transfer?
Google’s Magenta team has made Tone Transfer available for the public to try out Users can either choose a sample suggested by the program or even upload their tracks. Next, users have to choose the sound of the device in which the track will be morphed. Available result options are flute, saxophone, trumpet and violin.
First, users need to choose a source word and click the play icon below the word signature to get an idea of what it sounds like. After that, they need to choose an output sound and click play again to hear the sound created and converted to a new device.
Users will need to select the “Add Your Own” option under the Source Sample section if they want to upload their word samples. Users can either record a sound or upload a file from their PC if their device is compatible. Click the Done button after recording or uploading, this will allow you to listen to recorded or uploaded tracks (s). The Tone Transform tool will start processing the sample when you are happy with it and click the Transform button.
However, make sure you choose a track that plays only one tone at a time This will help the model to distinguish between different tones and create a better result. The tool also offers a slider icon that can be used to tune the final result. Users can adjust the mix, octave and height at their discretion.