As a winner of multiple awards, InfoTalk-Recognizer is widely accepted as the premier solution for applications that require multilingual, mixed-lingual automatic speech recognition (ASR) and/or speech-to-text (STT) and/or natural language understanding (NLU). The technology for multi-cultural environments such as trilingual Hong Kong, where Cantonese, Putonghua Chinese and English are commonly spoken.
The caller never picks a language
No menu, no prompt to choose one. The caller simply speaks, and Cantonese, Putonghua Chinese and English are recognised together within the same utterance.
Noise and accent tolerance
Stations, malls and service counters are loud, and the people speaking carry local accents. Recognition is engineered to hold its accuracy in those conditions, in city environments across Hong Kong.
One sentence, two languages.
唔好意思我想check一下我account嘅balance
幫我book張flight去東京下個禮拜三
我個card唔見咗要cancel佢
请帮我转到customer service谢谢
Deployed in Hong Kong call centres for two decades.
Ideal for these applications.
- Call centres
- Chatbots
- Voicebots
- Customer services
- Intelligent IVR systems
- Automated receptionists
- Speech analytics
- Transcription
- Voice control devices
- Voice typing
- Speech writing
- Audio-to-text conversion
7 facts about Cantonese and Chinese speech recognition
What is speech recognition?
Speech recognition is about converting the human voice into text. You speak, and the technology converts your spoken words into texts.
What is multilingual speech recognition?
In a multi-cultural multilingual environment like Hong Kong, people often mix English words with Cantonese or Chinese words when they speak. However, they speak these English words with unique non-native speaking accents, in contrast to native-speaking accents such as American English or British English.
Why is speech recognition challenging?
Everyone has a unique voice. Even when the same person speaks the same word several times, the acoustic voice signal would be different every time. Due to co-articulation, the acoustic signal of the word can also be affected by the words before and after it. Technologists call it a stochastic process, requiring sophisticated probabilistic modeling techniques. Environmental noise presents more hurdles for the technology to overcome.
Why is speech recognition in Cantonese and Chinese more challenging than in English?
Though Chinese is a tonal language, speech recognition in Putonghua is not necessarily more challenging than in English. However, Cantonese is more like a verbal language than a written language. The abundance of trendy idiosyncrasy, colloquialism and slang in spoken Cantonese can sometimes cause difficulties even to local people in Hong Kong.
What is the difference between artificial intelligence (AI) and speech recognition?
Speech recognition has always been a multi-disciplinary discipline involving computer science, artificial intelligence, signal processing, information theory, acoustics, language and linguistics, among others. With recent technology advances in artificial intelligence (AI), machine learning (ML) and deep neural networks (DNN), speech recognition has become more interdisciplinary than ever before.
Is speech recognition still in the laboratory stage?
Over the past few decades, speech recognition systems in Cantonese, Putonghua and English have been successfully deployed in a wide variety of industries in Hong Kong and Asia. Successful commercial deployment, especially in mission-critical high-traffic applications for public use, requires experience, expertise and knowhow that are above and beyond simplistic laboratory experiments.
Does speech recognition have other names?
There are other names for speech recognition, such as automatic speech recognition (ASR), speech-to-text (STT), voice recognition (VR), voice-to-text (VTT), 語音辨識, 文語轉換, 語音控制, 語音轉文字. They all mean more or less the same technology. However, Hong Kong speech recognition is unique in that people speaking Cantonese often mix it with English, so speech recognition in Hong Kong allows the speaker to mix Cantonese with English. The same applies to Putonghua speech recognition.
The product family
Built on the products
See it working on your own material.
Bring one workflow and a sample of your recordings. We strive to respond within 24 hours.
