A robot making clones of its voice - now quick and easy with tools like ElevenLabs.

You, But Fluent – Voice Cloning for Language Learners

I could barely contain my excitement in last week’s post on ElevenLabs’ brilliant text-to-speech voice collection. I’ve had a week of playing around with it now, and if anything, I’m only more enthusiastic about it.

After a bit of deep-delving, it’s the voice clone features that have me hooked right now. ElevenLabs can make a digital version of your voice from just 30 seconds of training speech. And it’s fast. I expected a bit of a wait for audio processing the first time I used it. But no – after reading in a couple of passages of sample text, my digital TTS voice was ready to use within seconds.

For a quick ‘n’ easy tool, it does a brilliant job of picking up general accent. It identified mine as British English, captured most of my Midlands features (it struggled with my really low u in bus, though – maybe more training would help), and it got my tone bang on. Scarily so… I can understand why cybersecurity pundits are slightly nervous about tech like this.

Your Voice, Another Language

The most marvellous thing, though, was using my voice to read foreign language texts. Although not 100% native-sounding – the voice was trained on me reading English, of course – it’s uncannily accurate. Listening to digital me reading German text, I’d say it sounds like a native-ish speaker. Perhaps someone who’s lived in Germany for a decade, and retains a bit of non-native in their speech.

But as far as models go, that’s a pretty high standard for any language learner.

ElevenLabs' TTS interface with the custom voice 'Richard' selected.

ElevenLabs’ TTS interface with the custom voice ‘Richard’ selected, ready to read some German.

The crux of it is that you can have your voice reading practice passages for memory training (think: island technique). There’s an amazing sense of personal connect that comes from that – that’s what you will sound like, when you’ve mastered this.

It also opens up the idea for tailoring digital resources with sound files read by ‘you’. Imagine a set of interactive language games for students, where the voice is their teacher’s. Incredible stuff.

In short, it’s well worth the fiver-a-month starter subscription to play around with it.

Open mic, ready for your voice

Voice in my head: developing polyglot personalities

If you learn more than one language, there is one question you hear more often than any other: how do you avoid getting mixed up? There are many answers to this. But one key strategy, for me, is developing a distinct voice for each language.

When you think about it, differentiating your languages by voice makes complete sense. Languages have patterns of pitch and tone quite distinct from one another. The voice you developed growing up with your native tongue adapted to fit the phonology of that language. It’s not surprising if the fit is a little less snug in French, Spanish, German and so on – at least without a little modification.

But changing your voice can feel intimidating. Our voice is a fundamental element of the self we project into the world; altering it can feel too bold, too cheeky. It’s no wonder that school students in the language classroom can feel reluctant to really get stuck into a foreign accent out. As adult learners, we face exactly the same fear. So how can we best approach a multiple voice approach to language learning?

Have fun with it

It’s important to remember that language learning regularly challenges us to act counter to our everyday inhibitions. Whether it’s speaking with strangers, supplementing our broken speech with frantic hand gestures, or trying to mimic how others sound, linguists are so often thrust out of reasonable, human comfort zones.

The best learners acknowledge this, and embrace the challenge head on. In short, it pays to be a clown!

It’s something we are naturals at as kids, chiefly because kids feel less embarrassment when they are playing around. For instance, this is one area where you needn’t feel guilty about wallowing in linguistic stereotypes. Have a hoot combining your French with dodgy Allo Allo accents. Watch back old episodes of Eldorado and have a go at your cheesiest Spanish. The important thing is to let go of your fear of sounding foolish through having fun.

It’s all very childish… So enjoy it!

Experiment with pitch

You can create instant results by simply experimenting with the pitch of your voice. Spanish and Russian feel more natural to speak when I lower my voice, for example – something my friends still find hilarious (even though I’m not deliberately trying to make them laugh!). (“Is that your Spanish voice again?” Yes. It is. I’m so glad you find it funny!)

On the other hand, I’m aware that the my voice is higher when I speak Norwegian – perhaps because this is easier with a tonal accent.

A lot of this is also to do with the voices you are exposed to as a learner. I listen to a lot of podcasts, for example. And through those, I hear all sorts of voices at all sorts of pitches. It has become a great, accessible way to ‘window shop’ for voices you are comfortable to use as a model. In fact, there are some well-documented techniques like shadowing, which use audio mimicry to drill your foreign language accent.

Exaggerate differences

Once you have a hook on the aural ‘feel’ of a language, you can focus on those differences as a means to differentiate. This is especially helpful if you study pairs, or sets of similar languages. These present a very particular set of advantages and challenges to learners.

I’ve studied both Polish and Russian on my language travels. As Slavic languages (albeit from different branches), they looked and sounded extremely similar to me as a beginner. But gradually, I found a way to mark the line between them through voice.

For instance, I sometimes find it easier to think of the sound of language in terms of shape. Through this lens, Russian is a language with quite sharp edges to my mind. On the other hand, the sound shapes of Polish seem much softer and curvier. When developing my Polish and Russian voices, I place a lot of weight on reflecting these demarcating characteristics.

(Disclaimer: after lots more study of both languages, I realise how different they can both be, now! Sorry for my previous ignorance, Polish and Russian natives! 🙂 )

Own your language

All this, of course, can greatly add to your sense of ownership over the foreign language. It’s vital to claim a language as your own, if you want a life-long relationship with it. And carving out a voice, even a personality, within it, will help you stake that claim.

It’s about feeling at home speaking it; saying “this is my German / Spanish / Uzbek” and being proud of the speaker you have created. Educational psychologists pore over methods to increase ownership in learning; as a language learner, voice work is a handy shortcut to do just that.

Voice in my head – split personality?

Finally, it’s interesting to see how this phenomenon plays out in bilinguals in the real world. Personally, I’ve found myself intrigued by polyglots who report a personality change when speaking another language. Through playing with voice and accent across my active languages, I think I can recognise that, too. Building up a distinct voice and personality in a language will inevitably create ‘another you’.

There is some research evidence to support this, too. However, the effect could be down to situational, rather than psychological factors (ie., bilinguals use different languages in different situations, and they would naturally act differently in those situations anyway – e.g., with colleagues rather than with family). Environmental or otherwise, though, it’s a fascinating thought, and one you can have a lot of fun with as a learner.

Developing a voice in your foreign languages goes hand in hand with perfecting an accent. Have fun playing around with it. And enjoy your polyglot personalities!