English translation is being prepared.
Matched fieldsSummary
←Search results
Search “Speech”
English translation is being prepared.
Matched fieldsSummary
StepFun releases the StepAudio 3 Music music generation foundation model
On September 16, 2026, StepFun released the StepAudio 3 series of audio models, which includes the music generation foundation model StepAudio 3 Music. According to official information, the series comprises 5 audio models in total, covering real-time speech, speech recognition, speech generation, audio generation, and music. The company also disclosed the performance of the other models in the series: Realtime reached 98.9% and 99.7% on Artificial Analysis's Conversational Dynamics and Speech Reasoning respectively, ranking first in both; ASR's word error rate is 1.7%, matching the best score on the leaderboard. The company states that the series can be used to build voice agents capable of handling interruptions, reasoning while speaking, and calling tools, and it is now available through the Voice AI Lab and blog, with the API also open. There are currently no subsequent development facts.
On September 16, 2026, StepFun released the StepAudio 3 series of audio models, which includes the music generation foundation model StepAudio 3 Music. According to official information, the series comprises 5 audio models in total, covering real-time speech, speech recognition, speech generation, audio generation, and music. The company also disclosed the performance of the other models in the series: Realtime reached 98.9% and 99.7% on Artificial Analysis's Conversational Dynamics and Speech R
Matched fieldsSummary
Google releases Gemini 3.8 Live audio model
Google released two real-time conversational audio models, Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, available starting today through the Gemini API, Google Workspace, and the Gemini app. According to Google DeepMind, Gemini 3.8 Live Extended Thinking ranks first on the Artificial Analysis Speech-to-Speech Quality Index with a score of 82.6, achieves a 68.6% task completion rate on τ-Voice agent tasks, and scores 97.7% on Big Bench Audio; Gemini 3.8 Live ranks second on the Speech Agent Arena. Both models support near-real-time visual input, automatic switching between 97 languages during a conversation, and can execute tool and API calls in the background without interrupting the conversation. On September 18, Google demonstrated a workshop assistant built with Gemini 3.8 Live Extended Thinking that can analyze the workspace, reason aloud, and guide the building process. On September 21, Google also demonstrated a programming tutoring Agent built with this model that can view the user's screen, reference the p5.js library through function calling, and explain code logic.
Google released two real-time conversational audio models, Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, available starting today through the Gemini API, Google Workspace, and the Gemini app. According to Google DeepMind, Gemini 3.8 Live Extended Thinking ranks first on the Artificial Analysis Speech-to-Speech Quality Index with a score of 82.6, achieves a 68.6% task completion rate on τ-Voice agent tasks, and scores 97.7% on Big Bench Audio; Gemini 3.8 Live ranks second on the Speech A
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
Qwen launches real-time simultaneous interpretation model Qwen3.8-LiveTranslate
On September 19, 2026, Qwen launched its new-generation real-time simultaneous interpretation model, Qwen3.8-LiveTranslate. According to the official release information, the model is based on the Interleave architecture and reduces average latency (LAAL) from 2.8s to 2.3s across 60 languages, while improving faithfulness, fluency, and conciseness. New capabilities include real-time speaker diarization, which can distinguish speakers in multi-party speech and preserve each person's voice through more stable voice cloning; bilingual synchronized display, presenting the source text and translation on the same screen; and long-context disambiguation, which uses conversation history to clarify names and terminology to ensure translation consistency. The model is now available on QwenCloud, and the official announcement also provides links to a blog post and case studies. There are currently no subsequent developments.
On September 19, 2026, Qwen launched its new-generation real-time simultaneous interpretation model, Qwen3.8-LiveTranslate. According to the official release information, the model is based on the Interleave architecture and reduces average latency (LAAL) from 2.8s to 2.3s across 60 languages, while improving faithfulness, fluency, and conciseness. New capabilities include real-time speaker diarization, which can distinguish speakers in multi-party speech and preserve each person's voice through
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary
Google Releases Gemini 4 Carbon Model, Focused on Long-Horizon Agentic Workflows
Google has released its flagship model Gemini 4 Carbon, built specifically for continuously running agentic workflows, with a focus on improving coding, agentic, scientific task, and full-stack web development capabilities. The model ranks first on the Artificial Analysis Intelligence Index v5.0 with a score of 59, surpassing Claude Opus 5.5, GPT-6 Astra, and GPT-6.1 Sol; the launch promotional price is $2 per million input tokens and $10 per million output tokens, with a 95% discount available for cached input tokens. Carbon adopts a phased rollout strategy, first opening to select cybersecurity defenders, then expanding the range of trusted partners through the Fairwind program, and afterward opening to developers, enterprises, and consumers.
Google has released its flagship model Gemini 4 Carbon, built specifically for continuously running agentic workflows, with a focus on improving coding, agentic, scientific task, and full-stack web development capabilities. The model ranks first on the Artificial Analysis Intelligence Index v5.0 with a score of 59, surpassing Claude Opus 5.5, GPT-6 Astra, and GPT-6.1 Sol; the launch promotional price is $2 per million input tokens and $10 per million output tokens, with a 95% discount available
Matched fieldsBody
Mistral Opens the Weights of Mistral Large 4, a 1-Trillion-Parameter Natively Multimodal Model
Mistral AI announced that it is opening the weights of its frontier model Mistral Large 4 (codename Le Chonk), which has 1 trillion parameters with 52 billion active parameters and native multimodal support. ML4 was trained in a European data center on 3,800 Nvidia Grace Blackwell GPUs, improving from 28% to 41% on Terminal-Bench 4.0, tying GLM-5.3 for the best coding-agent score while using an average of 47% fewer tokens per task. Open weights mean enterprises can deploy it on their own hardware and behind their firewalls, and the model offers three reasoning modes: low, medium, and high.
Mistral AI announced that it is opening the weights of its frontier model Mistral Large 4 (codename Le Chonk), which has 1 trillion parameters with 52 billion active parameters and native multimodal support. ML4 was trained in a European data center on 3,800 Nvidia Grace Blackwell GPUs, improving from 28% to 41% on Terminal-Bench 4.0, tying GLM-5.3 for the best coding-agent score while using an average of 47% fewer tokens per task. Open weights mean enterprises can deploy it on their own hardwar
Matched fieldsBody
Google releases Gemini 4 Carbon, topping 15 of 17 benchmarks
Google has released Gemini 4 Carbon, a frontier model aimed at cybersecurity and long-horizon coding tasks, topping 15 of Google's own 17 benchmarks, but it has not yet been made public and is available only to some cybersecurity defenders. It scores 59 on Artificial Analysis' Intelligence Index, surpassing Claude Opus 5.5 and up 9 points from Argon four weeks ago; on long-input reasoning it leads OpenAI Astra and Opus by 24 points. The introductory API price is $2 per million input tokens and $10 per million output tokens, half that of Opus 5.5, doubling after the discount period to Opus's rates. Google has not announced when it will broaden access.
Google has released Gemini 4 Carbon, a frontier model aimed at cybersecurity and long-horizon coding tasks, topping 15 of Google's own 17 benchmarks, but it has not yet been made public and is available only to some cybersecurity defenders. It scores 59 on Artificial Analysis' Intelligence Index, surpassing Claude Opus 5.5 and up 9 points from Argon four weeks ago; on long-input reasoning it leads OpenAI Astra and Opus by 24 points. The introductory API price is $2 per million input tokens and $
Matched fieldsBody
Nvidia Releases Nemotron 3.5 Transcribe Promptable Transcription Model
Nvidia has released the open-weight multilingual transcription model Nemotron 3.5 Transcribe, which ranks first on the Hugging Face Open ASR English open-source model leaderboard with an average word error rate of 5.33%. The model uses a 30B MoE architecture with 4B activated parameters, supports transcription in 40 languages and translation between 31 languages and English, can label speakers and distinguish up to 8 voices, handles recordings up to 1 hour long, but does not support real-time streaming. Users can provide instructions such as context and terminology before transcription; the model can be deployed with vLLM or run locally on a single GPU via NeMo-Speech.cpp; Nvidia has also released the data curation pipeline Granary v2 and a public manifest of approximately 463,062 hours across 48 languages.
Nvidia has released the open-weight multilingual transcription model Nemotron 3.5 Transcribe, which ranks first on the Hugging Face Open ASR English open-source model leaderboard with an average word error rate of 5.33%. The model uses a 30B MoE architecture with 4B activated parameters, supports transcription in 40 languages and translation between 31 languages and English, can label speakers and distinguish up to 8 voices, handles recordings up to 1 hour long, but does not support real-time st
Matched fieldsBody