FOR RECORD LABELS

Release Your Album Worldwide: Song Dubbing for Record Labels

FAMILIAR LABS RESEARCHALL ARTICLES

ABSTRACT

Song Dubbing is the first true song dubbing: an album carried into another language in the artist's own voice and performance, with nothing sung again. Familiar Labs works with record labels album by album, music videos included, every translated lyric approved by the artist.

Most of the world hears the sound, not the song

LUMINATE 2023
45.1%

of streams of 2023's top 10,000 tracks worldwide were songs not sung in English.

Luminate, 2023 year-end
40%

of US music listeners enjoy music in a language other than English.

Luminate, 2023 midyear

A hit travels as a sound: the voice, the beat, the hook. The lyric stays in one language, so most listeners abroad love the song without knowing what it says.

What a label gets

EVERY TRACK

The artist's own voice

The same recording: the phrasing, the breath, the emotion.

THE MUSIC VIDEOS

Dubbed with the album

The lips follow the new lyric, up to 4K.

THE MASTERS

Release-ready audio

A stereo mix in 24-bit PCM WAV with 144 dB of dynamic range, plus the voice track alone.

THE LANGUAGES

9 to start

English, French, German, Italian, Korean, Mandarin, Portuguese, Russian, Spanish.

The label picks the languages per album. Each is a new audience to release, promote and tour in.

How the work goes

Song Dubbing is white glove: our team works every album by hand with the label and the artist.

  1. 01

    The label sends the album.

    The tracks, the lyrics and the music videos, with the artist's consent.
  2. 02

    The lyric is translated.

    Against the melody: never fewer syllables than the original, the rhyme scheme reproduced, internal and multisyllabic rhymes carried where the language allows, the flow first, the genre matched, the meaning and intention kept.
  3. 03

    The artist approves every line.

    In every language, before anything is dubbed.
  4. 04

    The album is dubbed.

  5. 05

    The label releases it.

Why it is the first true song dubbing

Song DubbingTHE FIRST TRUE SONG DUBBING

Until now a song reached another language only by being sung again (by the artist, another singer or a new vocal made for it); here nothing is sung again, and the original performance carries over.

Before, songs were sung again: Lee Hyun recorded every language himself for MIDNATT's "Masquerade" in 2023, and Kevin Woo sang the Korean vocal of Lauv's "Love U Like That" before it was carried into Lauv's voice. Both opened the door.

Our Multimodal Localization Foundation Model (MLFM) was built to keep a person in every language, and a sung line asks the most of it.

The artist stays in control

++

The singer stays the singer.

It starts with the artist's consent and ends with the artist's approval: no line is released unapproved. Dubbing, not generating: no other singer stands in.

It is available to record labels today, album by album. Email the team with the album and the languages you have in mind.

QUESTIONS

What is Song Dubbing for record labels?

It carries a label's album and its music videos into other languages in the artist's own voice and performance, album by album, with every translated lyric approved by the artist.

Does the artist sing the album again?

No. The original recording carries over with the translated lyric; nothing is sung again.

Who approves the translated lyrics?

The artist or their team, line by line in every language, before anything is released. Where a line can land more than one way, the artist chooses.

How does a label start?

By email: Song Dubbing is a white-glove service, not self-serve. Send the Familiar Labs team the album and the languages you have in mind.

REFERENCES

  1. [1]Music Business Worldwide, Luminate 2023 year-end: non-English songs' share of the top 10,000 tracks
  2. [2]Associated Press via News4JAX, Luminate 2023 midyear report, July 12, 2023
  3. [3]The Korea Times, Hybe uses AI voice technology for Midnatt's multilingual single, May 15, 2023
  4. [4]NME, Lauv taps AI and a former K-pop idol to create Korean version of 'Love U Like That', 2023
  5. [5]Song Dubbing: the paper
  6. [6]The opposite of AI music generation
  7. [7]Our mission

Multimodal Translation

Familiar Labs builds generationally better dubbing with a Multimodal Localization Foundation Model (MLFM), the first foundation model for multimodal translation: one model decides what to localize across text, audio and visual, and shows you the choices. Your voice and lips, carried into new languages. They're a team of 7 with PhDs, backed by Samsung, RedBird, Gold House. Our mission