Following the recent release of Voice Search in Vietnamese on Androiddevices, Google has announced the arrival of Vietnamese Voice Search ondesktop computers via the Chrome browser.
When visiting google.com.vn on Chrome, users can click on the microphone icon to ask their question.
Amy Kunrojpanya, Head of Public Affairs & Communications for theGreater Mekong Sub-Region, Google Asia Pacific, said her company hopesthis will prove useful to Vietnamese consumers and help them unlock moreof the web more easily and intuitively.
According toKunrojpanya, each time Google brings Voice Search to a new language, itteaches computers to understand the sounds and words that make up spokenlanguage.
Google accomplished this by working with native speakers to collect speech samples to model the language, she said.
The Vietnamese-specific language model was built from the ground up.
"For Vietnamese, we worked with about 700 volunteers from universitiesin Hanoi and HCM City to collect about 480 hours of speech samples.Once we collected the samples, we were able to build acoustical andlanguage models that taught computers how to ‘recognise' Vietnamese,"she said.
Google spent two years working with the local volunteers.
"We were able to gather huge amounts of data from Google fans inVietnam, who were eager to help. Many people opened their doors to us tohelp the cause of making Vietnamese awesome," said Kunrojpanya.
She said the Vietnamese language had presented unique challenges. Themajor challenge was recognising tones and transcribing the diacriticscorrectly (for example, ca means to chant; ca, tomato; ca, fish).
Diversity of accents across Vietnam also required that Google widenthe sampling and double the amount of acoustic samples that are normallycollected for other languages.
Google tried to capture bothnorthern and southern accents, spending months "on lexicon developmenton a complicated language", she said.
"Voice Search canrecognise regional accents in Vietnamese, but it isn't 100 percentperfect. The good thing is that the language model improves as morepeople use it," she said.
Also, since in Vietnamese writing there is a space after each syllable, it is harder to know when a word begins and ends.
In contrast, in a language like English, whole words are separated by spaces, she said.
So Google introduced special handling of Vietnamese syllables so thatthey could be properly interpreted in the context of other syllablesaround them.
There were other challenges as well. For example,many Vietnamese Google users frequently leave out accents and tonemarkers when they search (for example, pho instead of pho).
"Sowe had to create a special algorithm to ensure accents and tones wererestored in the search results provided, and then our Vietnamese userswould see properly formatted text in the majority of cases," saidKunrojpanya.-VNA
When visiting google.com.vn on Chrome, users can click on the microphone icon to ask their question.
Amy Kunrojpanya, Head of Public Affairs & Communications for theGreater Mekong Sub-Region, Google Asia Pacific, said her company hopesthis will prove useful to Vietnamese consumers and help them unlock moreof the web more easily and intuitively.
According toKunrojpanya, each time Google brings Voice Search to a new language, itteaches computers to understand the sounds and words that make up spokenlanguage.
Google accomplished this by working with native speakers to collect speech samples to model the language, she said.
The Vietnamese-specific language model was built from the ground up.
"For Vietnamese, we worked with about 700 volunteers from universitiesin Hanoi and HCM City to collect about 480 hours of speech samples.Once we collected the samples, we were able to build acoustical andlanguage models that taught computers how to ‘recognise' Vietnamese,"she said.
Google spent two years working with the local volunteers.
"We were able to gather huge amounts of data from Google fans inVietnam, who were eager to help. Many people opened their doors to us tohelp the cause of making Vietnamese awesome," said Kunrojpanya.
She said the Vietnamese language had presented unique challenges. Themajor challenge was recognising tones and transcribing the diacriticscorrectly (for example, ca means to chant; ca, tomato; ca, fish).
Diversity of accents across Vietnam also required that Google widenthe sampling and double the amount of acoustic samples that are normallycollected for other languages.
Google tried to capture bothnorthern and southern accents, spending months "on lexicon developmenton a complicated language", she said.
"Voice Search canrecognise regional accents in Vietnamese, but it isn't 100 percentperfect. The good thing is that the language model improves as morepeople use it," she said.
Also, since in Vietnamese writing there is a space after each syllable, it is harder to know when a word begins and ends.
In contrast, in a language like English, whole words are separated by spaces, she said.
So Google introduced special handling of Vietnamese syllables so thatthey could be properly interpreted in the context of other syllablesaround them.
There were other challenges as well. For example,many Vietnamese Google users frequently leave out accents and tonemarkers when they search (for example, pho instead of pho).
"Sowe had to create a special algorithm to ensure accents and tones wererestored in the search results provided, and then our Vietnamese userswould see properly formatted text in the majority of cases," saidKunrojpanya.-VNA
