Fears raised over voice cloning
08 March 2018 18:22 GMT

Chinese tech firm Baidu is developing a "Deep Voice" solution that the company says can reproduce speaker identities with about 3.7 seconds of training data.

A believable, if low-quality, false voice can now be produced from a only single sentence of speech, reports Computing.

The company writes that ‘Voice cloning is expected to have significant applications in the direction of personalisation in human-machine interfaces', and there are naturally concerns about identity theft.

Tom Harwood, CPO and co-founder at voice security solutions provider Aeriandi, said to Computing:

"This technology is poised to transform personalisation in human-machine interfaces, but it raises serious concerns about voice biometric security systems. Soon, criminals will need just a few seconds of someone's voice to cheat a voice recognition security system - voice biometric authentication will be rendered useless.

Baidu has been working on Deep Voice for over a year, and had already managed to reproduce speaker identities with about half an hour of training data. With new developments, it has lowered that time to 3.7 seconds.

A believable, if low-quality, false voice can now be produced from a only single sentence of speech. Of course, more training leads to higher-quality results, especially if there is more than one sample to learn from.

Baidu has uploaded an array of system demos to Github, showing the capabilities of Deep Voice. These include voice cloning and various manipulations, such as changing the voice from male to female or British to American.

While the company writes that ‘Voice cloning is expected to have significant applications in the direction of personalisation in human-machine interfaces', there are naturally concerns about identity theft.

Tom Harwood, CPO and co-founder at voice security solutions provider Aeriandi, said:

"This technology is poised to transform personalisation in human-machine interfaces, but it raises serious concerns about voice biometric security systems. Soon, criminals will need just a few seconds of someone's voice to cheat a voice recognition security system - voice biometric authentication will be rendered useless.

Baidu has been working on Deep Voice for over a year, and had already managed to reproduce speaker identities with about half an hour of training data. With new developments, it has lowered that time to 3.7 seconds.

A believable, if low-quality, false voice can now be produced from a only single sentence of speech. Of course, more training leads to higher-quality results, especially if there is more than one sample to learn from.

Baidu has uploaded an array of system demos to Github, showing the capabilities of Deep Voice. These include voice cloning and various manipulations, such as changing the voice from male to female or British to American.

While the company writes that ‘Voice cloning is expected to have significant applications in the direction of personalisation in human-machine interfaces', there are naturally concerns about identity theft.

Tom Harwood, CPO and co-founder at voice security solutions provider Aeriandi, said:

"This technology is poised to transform personalisation in human-machine interfaces, but it raises serious concerns about voice biometric security systems. Soon, criminals will need just a few seconds of someone's voice to cheat a voice recognition security system - voice biometric authentication will be rendered useless.

Baidu has been working on Deep Voice for over a year, and had already managed to reproduce speaker identities with about half an hour of training data. With new developments, it has lowered that time to 3.7 seconds.

A believable, if low-quality, false voice can now be produced from a only single sentence of speech. Of course, more training leads to higher-quality results, especially if there is more than one sample to learn from.

Baidu has uploaded an array of system demos to Github, showing the capabilities of Deep Voice. These include voice cloning and various manipulations, such as changing the voice from male to female or British to American.

While the company writes that ‘Voice cloning is expected to have significant applications in the direction of personalisation in human-machine interfaces', there are naturally concerns about identity theft.

Tom Harwood, CPO and co-founder at voice security solutions provider Aeriandi, said:

"This technology is poised to transform personalisation in human-machine interfaces, but it raises serious concerns about voice biometric security systems. Soon, criminals will need just a few seconds of someone's voice to cheat a voice recognition security system - voice biometric authentication will be rendered useless.

Industry Events



Smart Security Week 24-26 Sep 18


BIOSIG 2018 26-28 Sep 18
ADAS 2018 Event supported by Planet Biometrics 26-28 Sep 18