Chennai Musicians and Producers Grapple With the Rise of AI in Music

Musicians and music producers in Chennai are increasingly grappling with the integration of artificial intelligence in music production as the technology begins rewriting the rules of the industry. From recreating legendary voices to launching AI-native virtual artists, the rapid adoption of AI tools has sparked a mix of experimentation and skepticism among local creators.

The shift comes as major players push the boundaries of the technology. Eros Innovation recently announced plans to bring back the voice of legendary singer Mohammed Rafi using AI, alongside launching seven AI-native artists. Additionally, the Collective Artists Network has launched Trilok, India's first AI rock band featuring no human musicians.

For local music producers, AI is already changing daily workflows. Joel Sakkari, a Chennai-based independent musician and producer who performs as Sakre, shared that clients frequently approach him with AI-generated tracks for film songs, jingles, or advertisements. Sakkari is tasked with proofreading the chord progressions and adding real guitar parts to legitimise the AI compositions.

Mack Raj, managing director of The Bangalore Studio, which uses AI tools like Google's Flow, Gemini's Omini, and Kling AI, noted that the technology significantly lowers production costs. While traditional song production can cost around Rs 50,000, AI audio production costs about Rs 10,000, and AI-generated videos cost approximately Rs 5,000 per minute. Despite the savings, Raj believes AI cannot replace human vocals or the soul of an artist.

Many established musicians remain highly skeptical of the technology's current capabilities. Vasu Dixit compared current AI music to early AI-generated images that featured rough edges and extra limbs, stating that the technology is still in its initial stages. Musician Ajith K Prakash, who performs as Ajoopan, uses AI to generate music videos but noted that the technology fails to capture the complex intricacies of instruments like the flute or violin.

There are also cultural concerns regarding indigenous music. Sivasubramanian Muthusamy, a member of the band Uru Paanar, explained that while AI might replicate the sound of traditional instruments like the parai or yazh, it cannot replicate the generations of practice, rituals, and memories associated with them.

Technical limitations also pose practical challenges for producers. Sakkari noted that AI-generated music often suffers from odd instrument tones and sudden changes in melody or voice. He recalled an instance where a client insisted he recreate an AI-generated chord that was physically impossible to play unless a musician had seven fingers.