TikTok embarks on AI music creation and machine learning hiring spree

The industry of AI-powered music creation apps is booming.

In May, music creation platform Splice, estimated to be worth nearly $500 million, launched an artificial intelligence-based music app called CoSo, which uses what Splice calls its “Complementary Sounds” artificial intelligence technology. ” to create music “in a fraction of a second”. .

Bandlab, meanwhile, the social music creation platform that recently raised $65 million, has an AI-powered app called SongStarter that it says can “generate royalty-free music in a few seconds”.

Could TikTok and its parent ByteDance be one of the next big players in the industry?

In July 2019, ByteDance acquired Jukedeck, a British AI Music startup specializing in creating royalty-free music for user-generated online videos.

This year, the company appears to be doubling down on its AI-powered music creation ambitions.

In May, ByteDance launched Mawf, a machine learning-based music creation app that analyzes incoming audio signals and then “plays back” those signals using what it says are machine learning models of musical instruments. music.

Moreover, according to a report by the news site Planet Tech, ByteDance has also launched a music-making app in China called “Sponge Band”.

According Technology planetThis music recording and editing app also includes AI tools and notes that it’s especially “useful for users who want to further enhance audio effects and background music for short videos.”

Now, according to a number of job postings spotted by MBW, TikTok and parent company ByteDance are hiring a number of highly skilled machine learning and AI music creation specialists in the United States and China.

Here are some of those roles:

Scientific researcher in musical creation (SAMI)

According to the job description for this role, which is based in Mountain View, Calif., this person will “lead cutting-edge research and development in music creation tools to lower the barrier of music creation for everyone.”

They will also be required to “transfer advanced technologies to TikTok products” and “explore new products with music intelligence technology at its core.”

Elsewhere in the job posting, TikTok states that “the mission of this team is to enable interaction and content creation using speech/audio/music-related technologies.”

The announcement continues: “We are focused on cutting-edge R&D in speech, audio and music processing. We’re looking for top talent to work on music creation tools, integrate them into various TikTok products, and ultimately bring joy to our global user base!

Music Machine Learning Research Engineer

This individual, who will be based in Seattle, according to the announcement, “will conduct cutting-edge research and development in music information retrieval, music creation, music/audio signal processing, interactive music and related fields.” .

Candidates should have a master’s or doctoral degree in computer science, mathematics, engineering, or a related field.

They will also need to have experience in one or more areas of “musical intelligence” and deep learning, including but not limited to:

  • Music information retrieval
  • AI music composition and creation
  • Digital Signal Processing for Music and Audio – Interactive Music


According to TikTok, this Seattle-based role will involve “cutting-edge research and development in multimodal (audio, vision, music, and text) machine learning models.”

TikTok says this person will need to have “experience with multimodal machine learning in speech and audio, computer vision, NLP, and music information retrieval.”

“The Voice Audio Music Team’s mission is to enable interaction and content creation using speech and audio-related technologies,” the announcement reads.

“We focus[ed] on cutting-edge R&D in areas such as speech and audio, music processing, natural language understanding and multimodal deep learning, we are looking for the best talent to work on these exciting technologies, integrate them in various TikTok products and ultimately bring joy to our worldwide userbase!”

Audio AI Software Engineer

According to the announcement (translated), this Beijing-based engineer will be responsible for “the research and development and business support of Bytedance’s audio content creation and consumption business scenarios.”

This person will need to have “a wealth of experience in digital signal processing and the development of artificial intelligence/deep learning systems”.

This experience should also be in the fields of “3D spatial audio rendering, audio pre-processing such as noise reduction/echo/de-reverberation, voiceprint/wake-up, sound event detection, speech recognition, natural language”.The music industry around the world

Sherry J. Basler