Anuj Diwan

@anuj_diwan

PhDing @UTCompSci | RS Intern @Meta Superintelligence Labs | Prev. @GoogleDeepMind, FAIR @AIatMeta, @AdobeResearch | 2021 BTech CSE @iitbombay. 🇮🇳🇺🇸

New York, NY

ajd12342.github.io

Joined May 2014

1,153 Following

980 Followers

Replies Reposts Media

Pinned
Anuj Diwan
@anuj_diwan
Jun 1
Excited to share that I’ve joined @Meta Superintelligence Labs in NYC as a Research Scientist Intern this summer, working with the Voice Team on speech LLMs and dialogue models. Looking forward to meeting NYC folks in speech, language, or AI. Email [email protected] or DM!
11
Anuj Diwan
@anuj_diwan
Sep 2
Played around with this a bit; color me impressed! Very cool to see my intern host and several other familiar names behind it. Congrats to the team!
AI at Meta
@AIatMeta
Sep 1
Introducing Muse Voice Transcribe, the first real-time audio perception model from Meta Superintelligence Labs. Muse Voice Transcribe delivers real-time streaming ASR, diarization with 20+ speakers, and endpointing. It’s multilingual with seamless code-switching and improves
Anuj Diwan
@anuj_diwan
Aug 21, 2025
ParaSpeechCaps has been accepted to the EMNLP 2025 Main Conference!
Anuj Diwan
@anuj_diwan
Mar 7, 2025
Introducing ParaSpeechCaps, our large-scale style captions dataset that enables rich, expressive control for text-to-speech models! Beyond basic pitch or speed controls, our models can generate speech that sounds "guttural", "scared", "whispered" and more; 59 style tags in total.
Anuj Diwan
@anuj_diwan
Mar 21, 2025
If you'd like an open-source text-to-speech model that follows your style instructions, consider using our ParaSpeechCaps-based model! Model: huggingface.co/ajd12342/parle… Paper: arxiv.org/abs/2503.04713
OpenAI Developers
@OpenAIDevs
Mar 20, 2025
Three new state-of-the-art audio models in the API: 🗣️ Two speech-to-text models—outperforming Whisper 💬 A new TTS model—you can instruct it *how* to speak 🤖 And the Agents SDK now supports audio, making it easy to build voice agents. Try TTS now at OpenAI.fm.
1
Anuj Diwan
@anuj_diwan
Mar 7, 2025
Introducing ParaSpeechCaps, our large-scale style captions dataset that enables rich, expressive control for text-to-speech models! Beyond basic pitch or speed controls, our models can generate speech that sounds "guttural", "scared", "whispered" and more; 59 style tags in total.
3

Anuj Diwan

Anuj Diwan’s posts