Adobe Launches Enhanced Speech: A Free AI Tool to Remove Background Noise and Improve Sound Quality

Adobe is a software company popular for its extensive range of products for creative professionals, such as photo editing software, video editing software, etc. About a year ago, Adobe released a web-based tool capable of recording and editing spoken audio clips called Project Shasta. Named after a city in Adobe in Northern California – Mount Shasta, the project aims to integrate Artificial Intelligence to enhance audio or podcasts. Project Shasta, later renamed Adobe Podcast, is inspired by the thought of letting people speak their hearts out in the form of podcasts. 

More recently, Adobe released a new tool in its already successful Adobe podcast project. This new tool, called Enhanced Speech, denoises audio and removes all unnecessary background noises. If you’ve ever recorded a podcast or general audio for other purposes, you would know that the final audio received is always full of noise and unwanted echo. Enhanced Speech not only denoises the audio but also facilitates its quality, making the recording no less than a studio-generated clip. With this new advancement, anyone can now conveniently create professional audio in the ordinary surrounding.

Adobe Podcast has mainly four impressive attributes –

  1. Trouble-free and easy editing of the audio
  2. Remote recording with easy synchronization
  3. Mic check feature
  4. Enhanced audio quality. 

The project uses Adobe Premier Pro’s speech-to-text technique to edit the audio. For recording with other people, it supports easy cloud sync of all the audio after recording locally. The AI-based mic-check feature gives the user an insight into whether he is too close to his microphone to analyze his recording. The fourth and most well-liked feature is for enhancing the quality of the audio. 

Speaking of this tool thoroughly, Enhanced Speech is basically an Artificial Intelligence powered software that refines and sharpens the frequencies of the uploaded audio for improvisation. Creating good-quality audio by visiting a studio takes a lot of time and effort and is costly. Enhanced Speech overcomes this major challenge by making the entire tool completely free of cost. The consumer must simply upload the audio file by drag and drop. The tool supports MP3 and WAV formats, and the user can upload an audio file of up to 1 GB or an hour-long period. It takes a few seconds to work on the audio; the next moment, the enhanced audio version is ready to download. One can try accessing the tool for free by signing in from their Adobe account and clicking on the link https://podcast.adobe.com/enhance

According to the tests conducted on this service by Ars Technica, the tool works the finest when the uploaded audio is free from any crosstalk or excessive noise. The team tested it on audio recorded from an IMac’s built-in default microphone with the person standing 10 feet away. The output audio was enhanced and seemed as if it was recorded with a professional microphone in a noise-free studio.

Unquestionably, Adobe Podcast, particularly the Enhanced speech tool, would become a go-to tool for every voice recording and all those who wish to share their heartfelt stories through a podcast. It is definitely a great innovation for creating professional recordings conveniently and in no time. 


Check out the Tool. All Credit For This Research Goes To Researchers on This Project. Also, don’t forget to join our Reddit page and discord channel, where we share the latest AI research news, cool AI projects, and more.

Tanya Malhotra is a final year undergrad from the University of Petroleum & Energy Studies, Dehradun, pursuing BTech in Computer Science Engineering with a specialization in Artificial Intelligence and Machine Learning.
She is a Data Science enthusiast with good analytical and critical thinking, along with an ardent interest in acquiring new skills, leading groups, and managing work in an organized manner.

🐝 Join the Fastest Growing AI Research Newsletter Read by Researchers from Google + NVIDIA + Meta + Stanford + MIT + Microsoft and many others...