Today, iFLYTEK officially launched its latest developed Spark simultaneous interpretation voice model, marking the debut of the first large model in China with end-to-end voice simultaneous interpretation capabilities. This innovative technology significantly improves the translation performance across various scenarios compared to iFLYTEK's previous translation technologies, while also greatly reducing the end-to-end response time.
The release of the Spark simultaneous interpretation voice model provides users with a smoother and more accurate interpretation experience. During the demonstration by iFLYTEK, the translation device equipped with the Spark simultaneous interpretation voice model achieved almost no delay in English to Chinese voice interpretation, making it ideal for use in scenarios such as international travel and exhibitions. This optimization not only significantly increases the speed of subtitle presentation but also ensures the accuracy of Chinese-English translation.
It is reported that the Spark simultaneous interpretation voice model supports reverse control of translation length, enabling real-time segmentation of ideas, contextual understanding, and information reorganization during the end-to-end voice-to-text translation process. Additionally, the streaming speech synthesis technology supports prosodic linking of ideas and adaptive speech rate adjustment, further enhancing the naturalness and fluency of translations.
In international communication scenarios, whether for casual conversations, business interactions, or high-difficulty industry translations, the Spark simultaneous interpretation voice model has demonstrated outstanding performance. Its completeness of content, accuracy of information, and language quality are all at the forefront of the industry, surpassing simultaneous interpretation technologies such as Google Gemini 2.0 and OpenAI GPT-4o. It can achieve simultaneous interpretation delays of less than 5 seconds, reaching the level of human expert interpreters.
The launch of iFLYTEK's Spark simultaneous interpretation voice model not only represents a significant breakthrough in domestic AI translation technology but also signifies a future where international communication will be more convenient and efficient.
