Skip navigation
Muse Voice Transcribe is MSL's first real-time audio perception model -- rolling out today.
0:00 / 3:03

Muse Voice Transcribe is MSL's first real-time audio perception model -- rolling out today.

Mark Zuckerberg

@finkd2.5M followers

Visit on X
5,470 likesDownloadOriginal post

1,130,187 views

Muse Voice Transcribe is MSL's first real-time audio perception model -- rolling out today. SOTA in streaming speech-to-text, it handles speaker diarization, and endpointing natively in a single model.

36 Comments

Comments from the original post on X (Twitter)

  • Mark Zuckerberg's profile picture

    @finkd

    The model decides when to listen. It waits a little longer on hard words and commits faster on easy ones, using adaptive delay to predict each token and increase accuracy. With adaptive delay, the model is near the pareto front on speed-accuracy tradeoff.

    664 likes

  • Mark Zuckerberg's profile picture

    @finkd

    It holds up on messy, real audio too -- trained across 70+ languages (with 25 validated at launch), handles mid-sentence code-switching, and manages hour-long sessions with 20+ speakers. You can bias it toward names and domain terms as well.

    387 likes

  • Mark Zuckerberg's profile picture

    @finkd

    Already powering dictation in the Meta desktop app and voice input in Muse Code. Live now on the Meta model API with a zero-data-retention tier.

    354 likes

  • Andi Marafioti's profile picture

    @andimarafioti

    @eliza_luth

    113 likes

  • Rohan Paul's profile picture

    @rohanpaul_ai

    wow, Meta is on fire. Congrats Mark. looks like a brilliant model.

    21 likes

  • Diligent Plane's profile picture

    @DiligentPlane

    mark is in grind mode rn Love to see it

    13 likes

  • Hervé "pyannote" Bredin's profile picture

    @hbredin

    Rule of thumb: If @pyannoteAI is missing from the benchmark, there's probably a reason. cc @finkd @AIatMeta @alex_conneau

    12 likes

  • …'s profile picture

    @logarithmatic

    Nice work; it seems there are a lot of American researchers there.

    11 likes

  • abhishek's profile picture

    @Abhishocked

    Nice to see so many releases from meta. Always rooting for meta. Btw What is the cost to transcribe an hour video? I have here explained why 0.1 usd per hour or less is the right price for transcription.

    9 likes

  • Maksim P. Martynyuk 🇺🇸🇺🇦's profile picture

    @Maksim_P_M

    Going to be amazing for the meta glasses

    6 likes

  • 𝕱𝖚𝖑𝖑 𝕶𝖊𝖑𝖑𝖞's profile picture

    @full_kelly_

    Incredible performance that blows the existing SOTA at detecting voices out of the water, but it's a shame you're not releasing an open weights model version

    4 likes

  • London Member Specialist Club's profile picture

    @foodchef247

    Is social media Algorithms now run by medieval dictators? Does your followers read your posts & replies? - ELQUOR

    4 likes

  • Allora's profile picture

    @AlloraNetwork

    The next constraint is still judgment. Once systems can hear continuously, they need intelligence that stays accurate as conditions change, not just a better transcript.

    4 likes

  • Alex Volkov's profile picture

    @altryne

    let's go! Just a week after Gemini 3.5 intelligence transcribe models! WIll put them to the test on @thursdai_pod 🫡 this week!

    4 likes

  • Elshayib's profile picture

    @elshayib_

    Bro no one cares Fable 5.1 around the corner fuck that

    3 likes

  • Freeman Jiang's profile picture

    @freemanjiangg

    Would love to see how this performs on

    3 likes

  • Ivan Fioravanti ᯅ's profile picture

    @ivanfioravanti

    Great release! Keep pushing!

    3 likes

  • utku's profile picture

    @utkuagi

    Congrats!

    3 likes

  • Jim Faster's profile picture

    @Jim_Faster

    Looks absolutely incredible. I wish it was open source but I understand this is a lot of work.

    3 likes

  • ExamAdda's profile picture

    @examaddaorg

    SOTA streaming STT with native diarization and adaptive delay is a serious flex well done.

    3 likes