←Search results

Search “Speech”

By relevance · Total 1347 results Clear search

Friday

  1. MarkTechPost

    Edge0 Releases Audio8-TTS-Realistic: A 0.6B Open-Source Voice Cloning TTS Model

    Edge0 has released the open-weight voice cloning TTS model Audio8-TTS-Realistic, with 601M parameters, supporting 23 languages, with weights and code released under Apache 2.0. The model requires only 1 reference audio clip and its corresponding text for zero-shot voice cloning, and can also directly generate speech when no reference voice is available. Self-reported results show an average error rate of 4.86% on the CV3 nine-language benchmark, lower than Qwen3-TTS-1.7B's 7.73%, but it lags behind 4B-class models on the MiniMax benchmark, with Thai WER reaching 48.57%.

    Edge0 has released the open-weight voice cloning TTS model Audio8-TTS-Realistic, with 601M parameters, supporting 23 languages, with weights and code released under Apache 2.0. The model requires only 1 reference audio clip and its corresponding text for zero-shot voice cloning, and can also directly generate speech when no reference voice is available. Self-reported results show an average error rate of 4.86% on the CV3 nine-language benchmark, lower than Qwen3-TTS-1.7B's 7.73%, but it lags beh

    Matched fieldsTitle、Summary、Body

    #ai-vertical#Model releases#Speech#Open-source ecosystem
  3. Rohan Paul

    Claude mobile upgrade: multi-account switching and CarPlay voice

    The Claude mobile app was updated this week: on iPad and iPhone Duo, generated content now opens beside the chat, and iOS and Android support logging in to multiple accounts at once with switching without logging out. On Chromebook, Googlebook, and foldables, an adjustable sidebar, trackpad and right-click support, and keyboard shortcuts have been added; voice mode on iOS and Android responds faster and recognizes speech better in noisy environments, and Apple CarPlay is supported on iPhone.

    Matched fieldsTitle、Summary、Body

    #ai-news#Product updates#Speech#English text is being prepared.
  4. Rohan Paul

    10-person team Phonic releases speech model Phonic-1, claims it can query and book in parallel during calls

    The 10-person startup Phonic has released a speech model, Phonic-1, which it says can execute queries and bookings before the caller has finished speaking. The model tops the Artificial Analysis speech-to-speech leaderboard with a score of 87.1, 0.2 points higher than xAI's Grok Voice 3.0, with a customer service task completion rate of 79.9% and a first-audio latency of 0.61 seconds. Phonic says it scored 90.2 on the Cekura call test set, but the public Cekura leaderboard lists Phonic v1 at 68.3%; the company has completed a $27.5 million Series A round led by First Round, bringing total funding to $31.5 million.

    The 10-person startup Phonic has released a speech model, Phonic-1, which it says can execute queries and bookings before the caller has finished speaking. The model tops the Artificial Analysis speech-to-speech leaderboard with a score of 87.1, 0.2 points higher than xAI's Grok Voice 3.0, with a customer service task completion rate of 79.9% and a first-audio latency of 0.61 seconds. Phonic says it scored 90.2 on the Cekura call test set, but the public Cekura leaderboard lists Phonic v1 at 68.

    Matched fieldsTitle、Summary、Body

    #ai-news#Model releases#Speech#Agents
  6. Elon Musk

    Neuralink's first product is named Telepathy, with claims that output bandwidth will far exceed speech and typing

    Neuralink's first product is called Telepathy, which can enable high-speed communication with a computer or other people through thought. Elon Musk stated that its output bandwidth will, within a few years, far exceed the fastest human speech or typing speed. The tweet cited a post about Neuralink engineers holding a 72-hour hackathon for a speech restoration research participant, rapidly prototyping multiple thought-communication features.

    Matched fieldsTitle、Summary、Body

    #ai-news#Industry news#Embodied AI#Product updates
  7. Artificial Analysis

    Phonic 1 Tops the Artificial Analysis Speech-to-Speech Index, Leading Grok with 87.1 Points…

    Artificial Analysis evaluations show that Phonic's Phonic 1 (High) ranks first among 22 models on the Speech to Speech Index with 87.1 points, leading Grok Voice 3.0's 86.9 points by a narrow margin. The model ranks first on the Tau Voice agent task with 79.9%, leading the second place by 5.4 percentage points; it ranks third on Big Bench Audio with 98.9%, and its Time to First Audio is 0.61 seconds. Phonic 1 is available through the Phonic API, with input audio costing $0.07 per minute and $4.20 per hour.

    Artificial Analysis evaluations show that Phonic's Phonic 1 (High) ranks first among 22 models on the Speech to Speech Index with 87.1 points, leading Grok Voice 3.0's 86.9 points by a narrow margin. The model ranks first on the Tau Voice agent task with 79.9%, leading the second place by 5.4 percentage points; it ranks third on Big Bench Audio with 98.9%, and its Time to First Audio is 0.61 seconds. Phonic 1 is available through the Phonic API, with input audio costing $0.07 per minute and $4.2

    Matched fieldsTitle、Body

    #editorial#Evaluations / benchmarks#Speech#Agents
  8. Unite.AI

    Can adding voice to AI tutors help students learn better?

    The arXiv preprint "When AI Tutors Speak" studied an online MBA finance course and found that voice tutors produced more conversational interaction, but weekly mastery showed no significant difference from text within the set range of ±0.3 standard deviations, and the cost of voice delivery was about 2.8 times that of text. The study was conducted by Shihao Yang, Marshall Van Alstyne, and Chrysanthos Dellarocas, with 86 students randomly assigned to groups; ultimately 29 tutored students participated in the modality analysis, and the results were limited by sample attrition and assessment methods.

    The arXiv preprint "When AI Tutors Speak" studied an online MBA finance course and found that voice tutors produced more conversational interaction, but weekly mastery showed no significant difference from text within the set range of ±0.3 standard deviations, and the cost of voice delivery was about 2.8 times that of text. The study was conducted by Shihao Yang, Marshall Van Alstyne, and Chrysanthos Dellarocas, with 86 students randomly assigned to groups; ultimately 29 tutored students partici

    Matched fieldsTitle、Summary、Body

    #hot-signal#Papers/Research#Speech#Tutorials / practice
  9. Unite.AI

    Rivian Licenses AI Voice Assistant Technology Through Volkswagen Joint Venture

    Rivian disclosed in its Q3 2026 earnings report that it has licensed its AI Voice Assistant technology to the Volkswagen Group through the Rivian and VW Group Technology joint venture, as part of the two companies' electrical architecture and software partnership. Rivian's software and services revenue for the quarter was $605 million, up 45% year-over-year, of which $354 million (59%) came from the joint venture. Rivian Assistant launched on the R2 in October 2026 and added Gmail integration via an OTA update.

    Rivian disclosed in its Q3 2026 earnings report that it has licensed its AI Voice Assistant technology to the Volkswagen Group through the Rivian and VW Group Technology joint venture, as part of the two companies' electrical architecture and software partnership. Rivian's software and services revenue for the quarter was $605 million, up 45% year-over-year, of which $354 million (59%) came from the joint venture. Rivian Assistant launched on the R2 in October 2026 and added Gmail integration vi

    Matched fieldsTitle、Body

    #hot-signal#Industry news#Speech#English text is being prepared.

Thursday

  1. Testing Catalog

    Rime Releases Speech Model Prism, Which Writes Text Before Speaking

    Rime has released Prism, a speech-to-speech model for AI voice agents that writes every reply as text before speaking. Enterprises can edit the transcript and block things the agent should not say, while Prism keeps the call uninterrupted and can look up accounts or freeze cards during the conversation. Rime says no other speech-to-speech model takes this approach, making Prism better suited to regulated enterprises such as banks.

    Matched fieldsTitle、Summary、Body

    #hot-signal#Model releases#Speech#Agents
  3. Testing CatalogSelected

    NVIDIA Releases Nemotron 3.5 Transcribe Open-Weight Speech Model

    NVIDIA has released Nemotron 3.5 Transcribe, an open-weight 30B MoE model with approximately 4B active parameters. The model supports transcription in 40 languages, speaker attribution, translation, contextual prompting, and audio understanding, and ranks first in VoiceArena's 14-system comparison of speaker diarization plus ASR.

    Matched fieldsTitle

    #hot-signal#Model releases#Speech#Open-source ecosystem
  5. Aravind Srinivas

    Aravind Srinivas: Voice is the natural form of using a computer

    Perplexity's Houssein Djirdeh announced that Perplexity's web version has launched a new voice mode, supporting natural conversation, full voice operation of the app, and multiple parallel sessions, based on GPT-Live-1. Perplexity CEO Aravind Srinivas reposted and commented, saying that voice is the natural form of using a computer.

    Matched fieldsTitle、Summary、Body

    #editorial#English text is being prepared.#Speech
  6. OpenBMB

    OpenBMB Open-Sources ESP32-S3 Voice Tower: MiniCPM5-2B + VoxCPM2

    OpenBMB ambassador @emwstudio has open-sourced an ESP32-S3 cockpit-style attitude indicator project that uses MiniCPM5-2B to generate tower replies and VoxCPM2 to synthesize speech. The device sends push-to-talk audio over Wi-Fi to a Mac-side pipeline, where Qwen3-ASR performs local recognition, and the reply audio then streams back to the device speaker. The tower incorporates simulated flight data such as altitude, speed, heading, pitch, and roll, and points out inconsistencies between requests and instrument readings.

    OpenBMB ambassador @emwstudio has open-sourced an ESP32-S3 cockpit-style attitude indicator project that uses MiniCPM5-2B to generate tower replies and VoxCPM2 to synthesize speech. The device sends push-to-talk audio over Wi-Fi to a Mac-side pipeline, where Qwen3-ASR performs local recognition, and the reply audio then streams back to the device speaker. The tower incorporates simulated flight data such as altitude, speed, heading, pitch, and roll, and points out inconsistencies between request

    Matched fieldsTitle、Summary、Body

    #editorial#English text is being prepared.#Speech#English text is being prepared.
  7. Artificial Analysis

    Fish Audio Drama 3 ranks first among open-weight models in two speech arenas, and first in pronunciation robustness

    Artificial Analysis evaluations show that Drama 3, released by Fish Audio, has an Elo of 1,278 in the Provider Voice scenario of Text to Speech (1,512 arena appearances), ranking 6th overall and 1st among open-weight models; in the Controlled Voice scenario, its Elo is 1,085 (1,958 appearances), ranking 12th overall and 1st among open-weight models. On pronunciation robustness, Drama 3 ranks 1st with 94.6% (2,157 rated clips), higher than the 93.2% of Ming-3.0-tts and Drama 3 Preview. Its generation speed is 66.6 characters per second, its pricing is $15/1M characters, and its weights are open under the Fish Audio Community License.

    Artificial Analysis evaluations show that Drama 3, released by Fish Audio, has an Elo of 1,278 in the Provider Voice scenario of Text to Speech (1,512 arena appearances), ranking 6th overall and 1st among open-weight models; in the Controlled Voice scenario, its Elo is 1,085 (1,958 appearances), ranking 12th overall and 1st among open-weight models. On pronunciation robustness, Drama 3 ranks 1st with 94.6% (2,157 rated clips), higher than the 93.2% of Ming-3.0-tts and Drama 3 Preview. Its genera

    Matched fieldsTitle、Body

    #editorial#Evaluations / benchmarks#Speech#Open-source ecosystem
  8. Analytics India Magazine

    NVIDIA Releases Nemotron 3.5 Transcribe Open-Weight Speech Model, Supporting 40 Languages

    On October 28, NVIDIA released Nemotron 3.5 Transcribe, an open-weight speech model that can transcribe audio in 40 languages, translate speech, and identify speakers, and developers can control the transcription format with natural language instructions without retraining. The model consists of a Transformer speech encoder, a speaker diarization model, and the Nemotron 3.5 Lightning language model, with approximately 4.3 billion activated parameters, and the released checkpoint is labeled 30B-A4B. NVIDIA says it achieves an average word error rate of 4.71% on nine public test sets of the English Open ASR Leaderboard and 4.62% on seven languages of the multilingual leaderboard, and calls it the best open-weight model on both leaderboards; the model is released in BF16, FP8, and NVFP4 formats as well as GGUF files, under the OpenMDW-1.1 license.

    On October 28, NVIDIA released Nemotron 3.5 Transcribe, an open-weight speech model that can transcribe audio in 40 languages, translate speech, and identify speakers, and developers can control the transcription format with natural language instructions without retraining. The model consists of a Transformer speech encoder, a speaker diarization model, and the Nemotron 3.5 Lightning language model, with approximately 4.3 billion activated parameters, and the released checkpoint is labeled 30B-A

    Matched fieldsTitle、Summary、Body

    #hot-signal#Model releases#Speech#Open-source ecosystem
  9. Analytics India Magazine

    Gnani AI Releases Prisma v3.0 Speech Recognition Model, Supporting 10 Languages

    Gnani AI has released its automatic speech recognition model Prisma v3.0, supporting nine Indian languages including Hindi and Bengali as well as English, and capable of transcribing multilingual conversations with intra-sentence Hindi-English code-mixing. On the Vistaar benchmark, the model achieves a macro-average WER of 16.93, lower than ElevenLabs Scribe v2's 19.98, Microsoft Azure Speech-to-Text's 20.51, and Deepgram Nova 3's 26.57. It offers four transcription modes—Verbatim, Transcribe, Code-mix, and Transliterate—and supports up to 200 custom vocabulary terms per request, raising the recall rate for these terms from 70.9% to 90.3%.

    Gnani AI has released its automatic speech recognition model Prisma v3.0, supporting nine Indian languages including Hindi and Bengali as well as English, and capable of transcribing multilingual conversations with intra-sentence Hindi-English code-mixing. On the Vistaar benchmark, the model achieves a macro-average WER of 16.93, lower than ElevenLabs Scribe v2's 19.98, Microsoft Azure Speech-to-Text's 20.51, and Deepgram Nova 3's 26.57. It offers four transcription modes—Verbatim, Transcribe, C

    Matched fieldsTitle、Summary、Body

    #hot-signal#Model releases#Speech#Multimodal
  10. gihyo.jp

    NVIDIA Releases Speech Transcription and Translation Model Nemotron 3.5 Transcribe

    On October 28, 2026, NVIDIA released the speech transcription and translation model Nemotron 3.5 Transcribe, supporting 40 languages including Japanese and capable of transcribing recordings up to 1 hour long. The model is an open-weight 30B MoE architecture that activates approximately 4B parameters during inference, can specify via text instructions whether to retain filler words and how numbers are written, and supports speaker diarization for up to 8 people and bidirectional Japanese-English translation. The weights are released on Hugging Face as NVIDIA-Nemotron-3.5-Transcribe-30B-A4B under the OpenMDW-1.1 license, with BF16, FP8, and NVFP4 versions and NeMo Speech fine-tuning tools provided.

    On October 28, 2026, NVIDIA released the speech transcription and translation model Nemotron 3.5 Transcribe, supporting 40 languages including Japanese and capable of transcribing recordings up to 1 hour long. The model is an open-weight 30B MoE architecture that activates approximately 4B parameters during inference, can specify via text instructions whether to retain filler words and how numbers are written, and supports speaker diarization for up to 8 people and bidirectional Japanese-English

    Matched fieldsTitle、Summary、Body

    #hot-signal#Model releases#Speech#Open-source ecosystem

Wednesday

  1. Artificial Analysis

    NVIDIA releases the Nemotron 3.5 Transcribe speech transcription model, tying for first among open-source models with 2.8% WER

    NVIDIA has released the open-weight speech transcription model Nemotron 3.5 Transcribe, with 30B total parameters and 4B activated parameters. On Artificial Analysis's non-streaming speech transcription leaderboard, it achieves a 2.8% AA-WER, tying for first among open-source models with Mistral's Voxtral Small. On the speech agent dialogue dataset AA-AgentTalk, it leads Voxtral Small's 2.3% with 1.8%. Its activated parameters are 4B, while Voxtral Small's are 24B; other open-source models of comparable scale have a WER of about 4%.

    NVIDIA has released the open-weight speech transcription model Nemotron 3.5 Transcribe, with 30B total parameters and 4B activated parameters. On Artificial Analysis's non-streaming speech transcription leaderboard, it achieves a 2.8% AA-WER, tying for first among open-source models with Mistral's Voxtral Small. On the speech agent dialogue dataset AA-AgentTalk, it leads Voxtral Small's 2.3% with 1.8%. Its activated parameters are 4B, while Voxtral Small's are 24B; other open-source models of co

    Matched fieldsTitle、Summary、Body

    #editorial#Evaluations / benchmarks#Speech#Open-source ecosystem