杏吧原创

It is easier to tell your own voice apart from other voices when listening to recordings through bone conduction headphones, because they mimic the vibrations from your skull that the brain detects as you speak.

Listening to a recording of your voice can sound very different from hearing it as you speak, because the self-voice you hear as you chat is filtered through your skull.

This was thought mainly to be due to the bone altering the auditory signal you perceive, but attempts to recreate self-voice by filtering the sound in a similar way to the skull have proved inconclusive.

Now, at the University of Geneva in Switzerland and his colleagues have replicated the experience by playing recordings through bone conduction headphones, which allowed people to better differentiate between their own voice and those of others.

To test how the bone conduction headphones changed people鈥檚 perceptions, Orepic and his team played 16 people short recordings of their own voice mixed in with a different voice from someone of the same gender in six different ratios, with from 15 to 85 per cent of the recording being their own voice.

Participants heard each mixture of voices 10 times in a random order for a total of 60 clips, through the bone conduction headphones or regular speakers, and were asked to say whether the voice more closely resembled their own or someone else鈥檚.

The researchers found that people could better distinguish their own voice when listening through bone conduction headphones, which pass vibrations straight to the cochlea in the inner ear, rather than relying on the vibrations that the sound causes in the eardrum to relay the information.

鈥淭he main conclusion is that we should change the way we think about self-voice,鈥� says Orepic. 鈥淚t鈥檚 not just an auditory stimulus, but it鈥檚 something multimodal.鈥� In natural scenarios, he says, there is always this component of vibrations that are picked up through the skin, for example.

Understanding the process better could be useful for helping live musicians who are listening to a recording of their own voice in their ear as they sing or to help people with schizophrenia who hear disembodied voices.

One idea about these voices is that they are created when the brain misattributes internal thoughts to external stimuli. 鈥淲e should first understand what it means to discriminate between voices, and then we could think of applications of maybe retraining people to re-recognise their own voice,鈥� says Orepic.

鈥淚t鈥檚 a very clever use of bone conduction headphones to precisely manipulate the bone conduction parameter and see how it affects perception,鈥� says at the University of Montreal in Canada.