What singing can teach you about copying an accent

Copying a note is a lot like copying an accent

In a study of 41 singers, how well they sang predicted how well they could imitate speech in an unfamiliar language, better than playing an instrument did (Christiner and Reiterer, 2013). A follow-up with 96 people found that singers beat both instrumentalists and non-musicians at imitating Hindi and English (Christiner and Reiterer, 2015).

The pattern holds across many studies. A 2025 meta-analysis of 57 studies and more than 3,000 people found that musical ability is reliably linked to learning a second language (Thompson, Salig and Slevc, 2025). You do not need to be a singer to sound natural. But the skills singers practice, careful listening and fine control of the voice, are the same ones accent practice builds.

The voice you copy changes how well you copy

This is where singing research gets interesting for accent learners. When scientists change only the model voice, accuracy changes with it.

• Your own voice is the easiest target. Adults matched pitches more accurately from a recording of their own voice than from another person’s voice or a plain tone (Moore and colleagues, 2008).
• A voice like yours comes next. Women matched pitch more accurately from a female voice than from a violin or clarinet. The researchers pointed to its sound being closer to their own (Watts and Hall, 2008).
• Children copy children best. Tested with 282 schoolchildren, a child’s voice got the most accurate answers, then a woman’s, then a man’s (Green, 1990).
• A different range pulls you off. Among 216 young singers who struggled to sing in tune, girls copied higher voices more accurately and boys copied lower ones. The model’s octave pulled their answers along with it (Price and colleagues, 1994).

If you have ever shadowed a teacher whose voice is much deeper or higher than yours and felt lost, this is the same effect. Your ear has to translate their voice into yours before you can even start. We wrote about that experience in why an accent is easier to copy in your own voice.

Hearing it is not the same as saying it

One finding runs through all of this. In the study above, people heard the pitches equally well whatever the voice; what improved with their own voice was singing them back (Moore and colleagues, 2008). Another study let people match recordings of their own voice either by moving a slider or by singing. They were far more accurate with the slider, which means the hard part was steering the voice, not hearing the note (Hutchins, Larrouy-Maestri and Peretz, 2014).

Practice is what closes that gap. People imitate recordings of themselves better than recordings of others, and researchers think part of the reason is that they have produced those patterns before. The more your voice has made a sound, the easier it is to make again (Pfordresher and Mantell, 2014).

Singing helps languages stick

Melody seems to help language stay with you. Adults who learned Hungarian phrases by singing them for 15 minutes remembered and reproduced them better than those who spoke them (Ludke, Ferreira and Overy, 2014). Children who learned an English passage by singing it pronounced its vowels better than children who spoke it, and the advantage was still there six months later (Good, Russo and Sullivan, 2015).

Spoken English has a melody of its own: the rise and fall of a question, the beat of stressed syllables, the small words that rush past between them. Listening for it the way you would listen to a tune is a good habit, whether or not you ever sing.

Bringing it to your accent practice

Put together, singing research suggests three habits for accent practice.

1. Copy a voice close to yours. The closer the model is to your own voice, the less translating your ear has to do. Younunciate takes this as far as it goes: each American English example is made in your own voice, from a short voice sample.
2. Listen for the melody. Rhythm, stress and linking carry a lot of what makes speech sound natural. Younunciate’s sound spellings show which syllables are strong, which shrink, and where words run together.
3. Say it, then hear yourself. Hearing is the easy half. Record yourself, compare it with the example, and try again. In Younunciate every take is kept, so you can hear how your voice changes, without a score telling you how to feel about it.