SAN FRANCISCO, July 28, 2026 /PRNewswire/ — OrcaRouter today announced the launch of OrcaDub, a frontier speech-to-speech video dubbing model designed to help creators, enterprises, and AI applications localize video content naturally across languages.
Unlike traditional dubbing workflows that translate text before synthesizing a new voice, OrcaDub performs end-to-end speech-to-speech translation while preserving the characteristics that make speech feel human—including voice identity, emotion, prosody, timing, and lip synchronization.
“As video becomes the dominant medium for communication, localization should feel native—not translated,” said Chris Zhang, MTS of OrcaRouter. “OrcaDub is built to preserve the original performance while making content accessible to audiences worldwide.”
Built for Natural Localization
OrcaDub is designed for production-quality multilingual video localization with a focus on naturalness and fidelity.
Key capabilities include:
- End-to-end speech-to-speech video dubbing
- Voice identity and speaker characteristic preservation
- Emotion and prosody transfer
- Idiom-aware translation and localization
- Song translation
- Per-word forced alignment
- Background music preservation
- OpenAI-compatible API for easy integration
The model enables developers to integrate high-quality video dubbing into creator tools, localization platforms, AI agents, media workflows, education products, customer support, and enterprise applications.
Early Evaluation Results
On OrcaDub’s multilingual evaluation suite, the model demonstrates strong performance across naturalness, translation quality, and speaker preservation.
Reported evaluation metrics include:
- 4.83/5 Mean Opinion Score (MOS)
- 96.8% Speaker Similarity
- 92.7 COMET Translation Quality
- 3.1% Word Error Rate (WER)
- 98.5% Speaker Attribution Accuracy
Developer-First API
OrcaDub is available through an OpenAI-compatible API, allowing developers to integrate multilingual dubbing with minimal changes to existing applications.
The platform is designed to support both synchronous and asynchronous video processing, enabling deployments ranging from short-form creator content to enterprise-scale localization pipelines.
Availability
OrcaDub is available today with transparent pay-as-you-go pricing at $0.60 per minute of processed video.
New users receive 10 free minutes to explore the platform. No subscription is required.
Learn more:
- Product: https://orcadub.orcarouter.ai
- Model card: https://orcadub.orcarouter.ai/model
- API Documentation: https://orcadub.orcarouter.ai/api-docs
About OrcaRouter
OrcaRouter is an AI infrastructure platform that provides intelligent model routing and developer APIs for production AI applications. Its products enable developers to access, optimize, and deploy frontier AI models through a unified interface. OrcaDub extends the platform into multilingual speech and video localization, bringing natural speech-to-speech translation to creators, enterprises, and AI-native applications worldwide.
Media Contact
OrcaRouter Communications
419580@email4pr.com
+1 650-609-7501
orcarouter.ai
View original content to download multimedia:https://www.prnewswire.com/news-releases/orcarouter-launches-orcadub-a-frontier-speech-to-speech-video-dubbing-model-for-natural-multilingual-localization-302835692.html
SOURCE OrcaRouter
