Home > News > Industry News > Voice translator "become popul.....


Voice translator "become popular" by strength

  • Author:Linki
  • Source:http://www.xinhuanet.com/info/
  • Release on:2019-09-11
Voice translator "become popular" by strength

Recently, the 4th Machine Translation Forum was held in Hangzhou. Participants shared the research and application results of machine translation frontiers, explored the opportunities and challenges of machine translation technology development, and the integration of artificial intelligence and translation has become a hot spot in the market.

The data shows that the current global language service market has great potential, and machine translation is a “red man” on the market. In 2018, the amount of online translation of machines is as high as 800 billion to 1 trillion words per day. At the same time, China's machine translation market demand is increasing day by day, mainly concentrated in enterprise users, involving multiple vertical areas such as petrochemical, electromechanical, transportation, finance, and tourism. “The traditional translation service model is labor-intensive, and machine translation has improved the efficiency in all aspects of the industry chain in the early stage,” said Tian Liang, CEO of the artificial translation machine service provider “New Translation Technology”.

Tian Liang believes that the current frontier applications of machine translation are mainly reflected in three aspects: First, the machine translation model is progressing rapidly, and the accuracy of the neural network-based translation model is increasing, which has brought professional translators at least 30% efficiency improvement; secondly, The concept of interactive machine translation has begun to be accepted by the industry, and the human-computer collaboration mode is accelerating. Third, there are more and more voice translation applications, from translation machines to translation headphones, to various types of intelligent robots. The speech translation mode has also been upgraded from the original speech recognition, machine translation, and speech synthesis "trilogy" to the "voice-to-speech" training mode.

“These three types of applications show the unique charm of the integration of artificial intelligence and translation industry. Machine translation is essentially a natural language processing technology. Technological advancement requires continuous research by industry and academia. This process will also be the language of artificial intelligence. The recognition ability is raised to a higher level." Tian Liang said.

However, some issues need to be resolved. Liu Wei, the chief operating officer of the classroom and the independent director of the DD translation officer, said: "The premise of many industries to connect artificial intelligence is to achieve standardization and dataization, but the language is difficult to standardize." Liu Wei said, in machine thinking, language The complexity and ambiguity make it difficult to achieve standardization and consistency. Therefore, the involvement of artificial intelligence in the translation industry is relatively simple, but it is not easy to do well.