StreamSpeech
Real-time speech translation, bridging cross-language communication.
PremiumNewProductProductivityReal-time translationMulti-task learning
StreamSpeech is a real-time speech-to-speech translation model based on multi-task learning. By learning translation and synchronization strategies in a unified framework, it effectively identifies the translation timing within streaming voice input, achieving a high-quality real-time communication experience. The model has demonstrated leading performance on the CVSS benchmark and can provide low-latency intermediate results, such as ASR or translation.
StreamSpeech Visit Over Time
Monthly Visits
No Data
Bounce Rate
No Data
Page per Visit
No Data
Visit Duration
No Data
StreamSpeech Visit Trend
No Visits Data
StreamSpeech Visit Geography
No Geography Data
StreamSpeech Traffic Sources
No Traffic Sources Data
