跳过导航
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

@finkd253万 位关注者

在 X 上查看
5,470 个赞下载原帖

1,130,187 次观看

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 条评论

评论来自 X (Twitter) 上的原帖

  • Mark Zuckerberg 的头像

    @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 个赞

  • Mark Zuckerberg 的头像

    @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 个赞

  • Mark Zuckerberg 的头像

    @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 个赞

  • Andi Marafioti 的头像

    @andimarafioti

    @eliza_luth

    113 个赞

  • Rohan Paul 的头像

    @rohanpaul_ai

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

    21 个赞

  • Diligent Plane 的头像

    @DiligentPlane

    mark is in grind mode rn Love to see it

    13 个赞

  • Hervé "pyannote" Bredin 的头像

    @hbredin

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

    12 个赞

  • … 的头像

    @logarithmatic

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

    11 个赞

  • abhishek 的头像

    @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 个赞

  • Maksim P. Martynyuk 🇺🇸🇺🇦 的头像

    @Maksim_P_M

    Going to be amazing for the meta glasses

    6 个赞

  • 𝕱𝖚𝖑𝖑 𝕶𝖊𝖑𝖑𝖞 的头像

    @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 个赞

  • London Member Specialist Club 的头像

    @foodchef247

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

    4 个赞

  • Allora 的头像

    @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 个赞

  • Alex Volkov 的头像

    @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 个赞

  • Elshayib 的头像

    @elshayib_

    Bro no one cares Fable 5.1 around the corner fuck that

    3 个赞

  • Freeman Jiang 的头像

    @freemanjiangg

    Would love to see how this performs on

    3 个赞

  • Ivan Fioravanti ᯅ 的头像

    @ivanfioravanti

    Great release! Keep pushing!

    3 个赞

  • utku 的头像

    @utkuagi

    Congrats!

    3 个赞

  • Jim Faster 的头像

    @Jim_Faster

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

    3 个赞

  • ExamAdda 的头像

    @examaddaorg

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

    3 个赞