Have you ever been in a position where someone is speaking nonstop for minutes, and you are struggling to make notes because it is just too fast to keep track? Well, though this was a significant issue in the past making stenography an art that required years of practice, the advent of technology may just have solved this problem for us. Innovative voice typing and speech recognition technology has now made it possible for us to convert speech into text almost on the go eliminating troubles we had earlier. Here’s a look at how speech recognition technology developed and why it’s so useful and also how we can make the best of it.
Appeal
Such technology entered the public domain quite recently, thanks to flashy press conferences held by tech giants claiming they have the best text to speech application out there. Sure to say, the appeal for it came naturally to us as humans love being understood and to think that a machine can now understand what we say seems fascinating to all of us.
History of Speech Recognition
- Audrey, one of the first voice recognition software created was the machine built by Bell Labs, and it understood digits with an accuracy of 90% which dropped to as low as 70% when someone other than the inventor spoke.
- Alexander Waibel worked on a machine developed at Carnegie Mellon which could spot and identify over 1000 words called Harpy.
- IBM Tangora, which was named after Albert Tangora who was at that point of time the fastest typist in the world, could adjust itself to the speaker’s voice, making it a massive improvement from previous machines.
- In 1997 the world’s first “continuous speech recogniser”, Dragon’s NaturallySpeaking software was released, and it was a landmark creation as it could identify 100 words per minute making it a doctor’s favourite.
- Machine learning, contributed heavily to this field allowing Google to develop its Voice Search App for the iPhone in 2008. The Hummingbird algorithm made the application more successful than all its predecessors, allowing it to be used on 50% of all smartphones. Then followed Siri, Cortana and Alexa, the three most popular speech recognition apps of today.
Greatest Obstacle
The greatest bane of such software is that every individual has a different voice and manner of speaking which makes standardisation very difficult. The spoken word varies substantially by dialects, speed, emphasis, gender and even level of education attained. Therefore, a significant obstacle faced by most applications that aim to convert word to text has been scaling or balancing this varied speech.
Present Scenario
Nearly half of Singapore has used voice typing services such as Siri, S Voice or Google Assistant at least once, and over a quarter use them every month. Advances in AI and machine learning has made a voice-activated world a reality that may be accomplished shortly. Most other developed and developing nations are also catching up with the percentage of voice typing app users increasing on a daily basis.
The average American speaks at a rate of 150 words per minute while a typist using a proper technique could clock around 40 words per minute and this is where voice recognition and typing software comes into the picture.
How to Set Up Speech-to-Text on Your Phone
These work for devices running Android 6.1 or higher and other devices will have similar steps with just a few variations.
- Go to Device Settings >Language and Input> Scroll to Speech section> Tap Voice input.
- Choose between:
- Basic Google recognition
- Enhanced Google services
- Then, press the Settings icon to manage your options.
- Tap on Languages and choose your preferred language.
- Tap Hands-free and choose whether you want the voice-typing feature on external microphones.
- Tap Offline speech recognition> Switch to All tab.
- Download the language packs you require.
How to Use
- Open any application that requires typing.
- Tap the microphone icon > Hold the Hold to talk button to start recording.
- Speak into the microphone, and let go when you’re finished.
- Tap the keyboard icon to get back to the standard keyboard.
Tips to Minimise Errors
- Speak clearly and don’t be too fast.
- Reduce background noises to the lowest possible to maintain clarity.
- Practice as the device gets better with time.
Benefits
- Helps you to take notes faster.
- Helps to free up your hands while driving.
- Makes the entire process more efficient.
- Minimises damage done to hands due to excessive typing.
Why must you make do with basic typing when you can dictate more efficiently? Thanks to developments in technology, you may not have to any longer as voice typing takes the world by a storm making the lives of note-takers around the world a whole lot easier. We may not be from a time wherein every command and action could be recognised and carried out by some application on our phones. While this will take some time to accomplish let us use the facilities available to us right now to become more efficient at note-taking.