videosdk-live

24 models • 1 total models in database

Sort by:

Namo Turn Detector V1 Multilingual

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Multilingual) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on mmBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in multilingual speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | ⚡ Latency | 📊 Evaluated on 25,000+ Multilingual utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1WEVVAzu1WHiucPRabnyPiWWc-OYvBMNj) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-English

license:apache-2.0

Namo-Turn-Detector-v1-Danish

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Danish) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Danish speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 86.52% | | 📈 F1-Score | 86.89% | | 🎪 Precision | 85.29% | | 🎭 Recall | 88.54% | | ⚡ Latency | > 📊 Evaluated on 700+ Danish utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Arabic

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Arabic) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Arabic speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 79.72% | | 📈 F1-Score | 82.70% | | 🎪 Precision | 72.97% | | 🎭 Recall | 95.42% | | ⚡ Latency | > 📊 Evaluated on 800+ Arabic utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Norwegian

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Norwegian) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Norwegian speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 87.34% | | 📈 F1-Score | 88.27% | | 🎪 Precision | 83.49% | | 🎭 Recall | 93.63% | | ⚡ Latency | > 📊 Evaluated on 1500+ Norwegian utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Vietnamese

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Vietnamese) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Vietnamese speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 82.37% | | 📈 F1-Score | 83.44% | | 🎪 Precision | 78.24% | | 🎭 Recall | 89.37% | | ⚡ Latency | > 📊 Evaluated on 1000+ Vietnamese utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Spanish

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Spanish) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Spanish speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 86.71% | | 📈 F1-Score | 87.81% | | 🎪 Precision | 78.98% | | 🎭 Recall | 98.88% | | ⚡ Latency | > 📊 Evaluated on 1200+ Spanish utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Dutch

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Dutch) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Dutch speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 90.00% | | 📈 F1-Score | 90.87% | | 🎪 Precision | 86.15% | | 🎭 Recall | 96.13% | | ⚡ Latency | > 📊 Evaluated on 800+ Dutch utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-German

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-German) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in German speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 91.91% | | 📈 F1-Score | 92.21% | | 🎪 Precision | 88.54% | | 🎭 Recall | 96.20% | | ⚡ Latency | > 📊 Evaluated on 1000+ German utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Russian

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Russian) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Russian speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 84.14% | | 📈 F1-Score | 85.64% | | 🎪 Precision | 81.66% | | 🎭 Recall | 90.02% | | ⚡ Latency | > 📊 Evaluated on 1400+ Russian utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Ukrainian

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Ukrainian) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Ukrainian speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 86.22% | | 📈 F1-Score | 85.80% | | 🎪 Precision | 83.04% | | 🎭 Recall | 88.76% | | ⚡ Latency | > 📊 Evaluated on 900+ Ukrainian utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Hindi

license:apache-2.0

Namo-Turn-Detector-v1-Finnish

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Finnish) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Finnish speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 84.75% | | 📈 F1-Score | 86.27% | | 🎪 Precision | 78.19% | | 🎭 Recall | 96.22% | | ⚡ Latency | > 📊 Evaluated on 1000+ Finnish utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Italian

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Italian) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Italian speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 86.82% | | 📈 F1-Score | 88.09% | | 🎪 Precision | 80.04% | | 🎭 Recall | 97.94% | | ⚡ Latency | > 📊 Evaluated on 700+ Italian utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Polish

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Polish) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Polish speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 87.80% | | 📈 F1-Score | 87.94% | | 🎪 Precision | 82.82% | | 🎭 Recall | 93.73% | | ⚡ Latency | > 📊 Evaluated on 900+ Polish utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Portuguese

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Portuguese) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Portuguese speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 86.90% | | 📈 F1-Score | 87.95% | | 🎪 Precision | 79.42% | | 🎭 Recall | 98.52% | | ⚡ Latency | > 📊 Evaluated on 1300+ Portuguese utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Turkish

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Turkish) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Turkish speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 96.79% | | 📈 F1-Score | 96.73% | | 🎪 Precision | 97.04% | | 🎭 Recall | 96.42% | | ⚡ Latency | > 📊 Evaluated on 900+ Turkish utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Indonesian

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Indonesian) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Indonesian speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 87.12% | | 📈 F1-Score | 87.78% | | 🎪 Precision | 82.38% | | 🎭 Recall | 93.93% | | ⚡ Latency | > 📊 Evaluated on 900+ Indonesian utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Bengali

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Bengali) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Bengali speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 79.20% | | 📈 F1-Score | 78.94% | | 🎪 Precision | 78.31% | | 🎭 Recall | 79.59% | | ⚡ Latency | > 📊 Evaluated on 900+ Bengali utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-French

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-French) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in French speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 84.99% | | 📈 F1-Score | 86.96% | | 🎪 Precision | 78.17% | | 🎭 Recall | 97.96% | | ⚡ Latency | > 📊 Evaluated on 1200+ French utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Marathi

license:apache-2.0

Namo-Turn-Detector-v1-Chinese

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Chinese) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Chinese speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 88.78% | | 📈 F1-Score | 89.78% | | 🎪 Precision | 84.26% | | 🎭 Recall | 96.08% | | ⚡ Latency | > 📊 Evaluated on 900+ Chinese utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Korean

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Korean) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Korean speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 97.30% | | 📈 F1-Score | 97.32% | | 🎪 Precision | 96.46% | | 🎭 Recall | 98.19% | | ⚡ Latency | > 📊 Evaluated on 800+ Korean utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0

Namo-Turn-Detector-v1-Japanese

[](https://opensource.org/licenses/Apache-2.0) [](https://onnx.ai/) [](https://huggingface.co/videosdk-live/Namo-Turn-Detector-v1-Japanese) ![Inference Speed - 🔄 Incomplete utterances (user will continue speaking) Built on DistilBERT architecture and optimized with quantized ONNX format, it delivers enterprise-grade performance with minimal latency. - Turn Detection Specialist: Detects end-of-turn vs. continuation in Japanese speech transcripts. - Low Latency: Optimized with quantized ONNX for | Metric | Score | |--------|-------| | 🎯 Accuracy | 93.52% | | 📈 F1-Score | 93.87% | | 🎪 Precision | 89.61% | | 🎭 Recall | 98.57% | | ⚡ Latency | > 📊 Evaluated on 800+ Japanese utterances from diverse conversational contexts [](https://colab.research.google.com/drive/1DqSUYfcya0r2iAEZB9fS4mfrennubduV) [](https://colab.research.google.com/drive/19ZOlNoHS2WLX2V4r5r492tsCUnYLXnQR) To use this model, you will need to install the following libraries. You can run inference directly from Hugging Face repository. Integrate this turn detector directly with VideoSDK Agents for production-ready conversational AI applications. > 📚 Complete Integration Guide - Learn how to use `NamoTurnDetectorV1` with VideoSDK Agents This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

license:apache-2.0