Most Viewed Content:

Jim Keller: Innovation is happening, RISC-V will win

In the global market, x86 architecture in the high-performance...

Tesla executives: Model 3 / Y is too good to sell

Martin Viecha, head of investor relations for Tesla, was...

Porsche Taycan electric car crashes and catches fire, driver burned to death

Porsche's electric car Taycan has recently been caught in...

OpenAI Announces Open Source Multilingual Speech Recognition System Whisper

Although tech giants, including Google, Amazon and Meta, have each put their own powerful speech recognition systems at the heart of their software and services. But speech recognition remains a challenging topic in artificial intelligence and machine learning. The good news is that today OpenAI is solemnly announcing the open-source of Whisper – known as an automatic speech recognition system that officially claims that it can perform powerful transcriptions in multiple languages and translate them into English.

(Source: OpenAI Blog)

What makes Whisper different, OpenAI says, is that it took 680,000 hours of multilingual and “multitasking” training data collected from the network, improving the scheme’s ability to recognize unique accents, background noise and technical terms.

The overview on the official GitHub repository says:

"The primary target users of Whisper models are AI researchers who study the robustness, generalization, capability, bias, and constraints of current models.

At the same time, it is also suitable as an automatic speech recognition solution for developers, especially English speech recognition.

Interested friends can download multiple versions of the Whisper system from the hosting platform, and its models show strong ASR results in about 10 languages.

In addition, if fine-tuned on certain tasks, they are also expected to show additional capabilities in application scenarios such as voice activity detection and narrator classification."

Architecture

Unfortunately, Whisper has not been robustly evaluated in related fields, and the model has its limitations – in the field of text prediction.

Since the system was trained on a lot of “noisy” data, OpenAI decided to give everyone a shot in advance, warning that Whisper could include words in the transcription that weren’t actually spoken.

The reason may be that Whisper is both trying to predict the next word in the audio and trying to transcribe the audio itself.

Process example

In addition, Whisper’s performance in different language scenarios is also inconsistent, especially when it comes to narrators of languages that are not well represented in the training data, the error rate is also higher.

But the latter is nothing new in the field of speech recognition, and even the industry’s premier systems have been plagued by such biases.

Referring to the results of a study shared by Stanford University in 2020 – systems from Amazon, Apple, Google, IBM and Microsoft have a much lower error rate (about 35%) for white users compared to blacks.

About 1/3 of Whisper’s audio dataset is non-English

Even so, OpenAI believes that Whisper’s transcription capabilities can be used to improve existing accessibility tools. It wrote on GitHub:

"While the Whisper model is not suitable for real-time transcription out of the box, its speed and size suggest that others can build on it for near real-time speech recognition and translation applications.

Beneficial applications built on top of Whisper's models, whose value is a tangible indication of the different capabilities of these models, are expected to have real economic impact.

We hope that everyone will actively use this technology for beneficial purposes, making improvements to automatic speech recognition technology easier and enabling more participants to create more responsible projects.

With the dual benefits of speed and accuracy, Whisper will allow for an affordable automated transcription and translation experience for large volumes of communications."

Latest

iOS 16.1 Beta 2 optimizes power icons: power slots follow actual power changes

Apple has redesigned the battery icon in iOS 16...

Tesla launches CCS adapter in the U.S. to charge at other charging posts for $250

Tesla recently launched the CCS adapter in the US...

ZTE Axon 30S will launch on September 26 featuring an under-display selfie camera sensor

On September 26 the ZTE Axon 30S will launch with an under-display selfie camera sensor giving the device a complete full screen display.

Microsoft Surface Studio 3 new product through the U.S. FCC certification

According to some online revelations and FCC documents, Microsoft...
spot_img

Newsletter

Don't miss

iOS 16.1 Beta 2 optimizes power icons: power slots follow actual power changes

Apple has redesigned the battery icon in iOS 16...

Tesla launches CCS adapter in the U.S. to charge at other charging posts for $250

Tesla recently launched the CCS adapter in the US...

ZTE Axon 30S will launch on September 26 featuring an under-display selfie camera sensor

On September 26 the ZTE Axon 30S will launch with an under-display selfie camera sensor giving the device a complete full screen display.

Microsoft Surface Studio 3 new product through the U.S. FCC certification

According to some online revelations and FCC documents, Microsoft...

More than 50% of cars sold in the U.S. are expected to be electric by 2030

According to foreign media reports, Tesla's headquarters in the...
Threza Gabriel
Threza Gabrielhttps://www.techgoing.com
TechGoing is a global tech media to brings you the latest technology stories, including smartphones, electric vehicles, smart home devices, gaming, wearable gadgets, and all tech trending.
spot_imgspot_img

Xiaomi 12T Pro will have a 200-megapixels camera; the price and specs leaked

Xiaomi will be soon launching its Xiaomi 12T and Xiaomi 12T Pro smartphones in the global markets. The Xiaomi 12T Pro is expected to...

Google Chrome feature for Android: fingerprint verification required to access Incognito tabs

Starting in 2020, a number of Google apps for iOS have added a Privacy Screen feature that requires biometric authentication via Touch ID or...

Google is sued in India for real money games: alleged unfair competition

According to reports, Indian online gaming platform Winzo today filed a lawsuit against Google to prevent Google from providing "real-money games" (real-money games) services....