←Speech
← All topics

Speech

Speech recognition, synthesis, real-time conversation and audio models.

Public reading groups
77
Last 30 days
17
Directory status
Curated topic

LatestApple Intelligence will enter China in December, with Alibaba's Qwen powering Siri and other features

Latest reading groups

Duplicate reports of the same fact are grouped

  1. X.PIN
    AI score75

    Apple Intelligence will enter China in December, with Alibaba's Qwen powering Siri and other features

    The first batch of Apple Intelligence features will enter China in December, with Alibaba's Qwen providing answers in Siri, Writing Tools, and Visual Intelligence. Qwen is opt-in, and Alibaba says it will not use requests and replies to train models, and will seek consent before sending any question, document, or image.

    Why read thisThis news specifies the timing of Apple Intelligence's entry into China, the partner involved, and how privacy is handled, which can be used to assess the partnership path for localizing overseas models.

  1. Hugging FaceFirst-partyModel releases
    AI score60

    NVIDIA Releases Nemotron 3.5 Transcribe and Granary v2 Data Pipeline

    NVIDIA has released the open-weight model Nemotron 3.5 Transcribe on Hugging Face and ModelScope, integrating transcription, speaker diarization, translation, and audio question answering into a single checkpoint. It supports transcription in 40 languages, translation between 31 languages and English, and speaker labeling in 13 languages, can process recordings up to 1 hour long, and can label up to 8 speakers. The model ranks first among open-weight models on the English leaderboard of the Hugging Face Open ASR Leaderboard with an average WER of 5.33%, and achieves 2.8% AA-WER on Artificial Analysis with approximately 4B activated parameters. The simultaneously released Granary v2 covers 48 languages and 463,062 hours of speech data, and the accompanying NeMo Speech fine-tuning recipe reduced WER on held-out speakers from 22.44% to 12.17% in a medical case.

    Why read thisThe original text provides unified weights for multilingual transcription, speaker diarization, and translation, along with public leaderboards and fine-tuning recipes, making it easy to assess deployment approaches.

  1. Google BlogFirst-partyModel releases
    AI score78

    Google DeepMind releases world model Genie 4, supporting real-time interactive simulation and native audio

    Google DeepMind releases the world model Genie 4, focused on real-time interactive simulation and introducing native audio for the first time in the series. The model is built on capabilities from the Gemini family: Gemini Omni is responsible for making simulations match the appearance and behavior of the real world, while other Gemini models provide the complex reasoning needed to maintain simulation coherence. Users can control scenes in real time using natural language, such as moving the camera, changing the weather, or introducing new characters, and each world also generates ambient sound, music, and interactive dialogue on the fly.

    Why read thisThe original text explains that Genie 4 adds native audio and natural language control to real-time interactive simulation, allowing readers to understand the new capability boundaries of world models.

  1. Ollama BlogFirst-partyProduct release
    AI score71

    Ollama Launches Speech Model Phonon, Supporting Text-to-Speech

    Ollama can now convert text to speech. Its first speech model is the Phonon series of small text-to-speech models from Gradium. It is currently available on macOS, with Windows and Linux support coming soon. On a MacBook Pro equipped with M5 Max, Ollama, with the help of MLX acceleration, is 35% faster than Gradium's own runtime, and its word error rate is on par with the Hugging Face Transformers reference implementation. Phonon is compatible with OpenAI's /v1/audio/speech interface, so existing OpenAI SDK code only needs to change the base URL; phonon is released under Apache 2.0, and the 40M model uses the CC BY-NC 4.0 license and is for non-commercial use only.

    Why read thisOllama has launched a speech model and made it compatible with the OpenAI API, so readers can use this to assess the integration cost and platform limitations of local TTS.

  1. Unite.AIModel releases
    AI score78

    Block releases full-duplex speech model Dessa, trained from scratch using only human conversation data

    Block's internal AI research lab released the full-duplex speech-to-speech model Dessa on October 23, stating that it was trained entirely from scratch on human conversational speech, can listen and speak simultaneously, and natively perceives emotion, sarcasm, and laughter. The model's inference speed is faster than real time, and it can run locally on a MacBook Pro M4 Max or higher configuration; the team completed training with a single NVIDIA GB200 NVL72. Block has published demo recordings and a public waitlist, and has opened early access to a small number of AI safety researchers.

    Why read thisDessa was trained from scratch using only speech data, demonstrating how full-duplex conversation handles emotion and interruption timing, and also providing local running requirements.

  1. ElevenLabs BlogFirst-partyProduct release
    AI score60

    ElevenLabs launches Eleven Realtime Translation real-time speech translation model

    ElevenLabs releases the real-time speech translation model Eleven Realtime Translation, currently available in alpha form to ElevenAPI enterprise customers. The model uses the low-latency speech model Eleven v4 Turbo by default, supports more than 90 languages and automatic language detection, keeps the same voice consistent across languages, and can be selected from a library of more than 17,500 voices or custom voice clones. Teams can adjust the translation startup speed, trading off between high-quality translation and low latency on complex calls; Deutsche Telekom has already built new call features based on ElevenLabs in its Magenta AI Call Assistant.

    Smiling person with tousled hair against a neutral background.

    Why read thisThe original text explains the adjustable-latency design and multilingual consistency of the real-time translation model, allowing readers to judge how it would be implemented in customer service and telecom scenarios.

  2. mistral_aiFirst-partyModel releases
    AI score80

    English translation is being prepared.

    back
  3. elevenlabs_blogFirst-partyModel releases
    AI score70

    English translation is being prepared.

    Benchmark table shows v1 Max outperforming Mini on speech reasoning and task success.

  1. Ant LingProduct release
    AI score67

    Ant Ling's Ming-3.0 Speech Models Get Day-0 Support in vLLM-Omni

    Ant Ling has open-sourced 5 speech models covering TTS, audio understanding, and ASR, with total parameter counts ranging from 0.5B to 126B (6.5B active parameters). vLLM announced that vLLM-Omni provided Ming-3.0 support on the day of release; Ant Ling thanked them for this and said deployment is now more convenient.

    Why read thisvLLM-Omni provided support on the very day Ming-3.0 was open-sourced, giving readers a way to understand the deployment path for this batch of speech models.

  2. inclusionai_new_reposFirst-partyModel releases
    AI score69

    English translation is being prepared.

    Ming-3.0-tts Logo

  1. GeminiProduct release
    AI score67

    Gemini 4 Argon can turn video or camera footage into a playable synthesizer

    Google Gemini announced that Gemini 4 Argon can reimagine video or camera footage as a playable synthesizer. Users can quickly build interactive audio applications in Canvas, set the mood, and use the camera as an instrument to play.

    Why read thisGoogle Gemini officially demonstrated how to turn camera footage into a playable synthesizer in Canvas, so readers can learn about its new multimodal interaction use cases.

  2. apple_newsroomFirst-partyProduct release
    AI score67

    English translation is being prepared.