Shunya LabsShunya LabsPlayground

Radiance · Voice Cloning

Voice Cloning

Create a reusable AI voice from a short reference recording for voice agents, multilingual applications, narration, customer experiences, media, and personalised speech. Provide a clean voice sample, enter new text, and use Shunya Radiance to generate speech that follows the reference speaker's voice characteristics - including across supported languages.

3-6 sec reference audioCross-lingual synthesis11 expression stylesAPI inline cloning
Reference sample

Welcome to Shunya. This is my reference voice sample.

3-6 seconds of clean speech
New speech · Radiance

Your appointment has been confirmed for tomorrow morning at ten.

Generated from new text
Reference · 3-6 secCross-lingualExpression controlPOST /v1/audio/speech

Voice Cloning

A few seconds of voice. New speech from New Text.

Traditional voice recording requires the speaker to record every sentence that an application may eventually need. Voice cloning changes that workflow. Provide a short reference recording and Radiance can use that reference to generate entirely new speech.

3-6 sec

Recommended reference audio for voice cloning

1

Reference voice used to generate new sentences

Cross-lingual

Cloned voices can synthesize supported languages

11

Expression styles available through the Radiance speech stack

CapabilityRadiance Voice Cloning
Voice Cloning✓
Short Reference Audio3-6 sec
New Text Generation✓
Cross-lingual Synthesis✓
Expression Control✓
Speed Control✓
Batch Generation✓
Streaming TTS✓
Multiple Audio Formats✓
API Integration✓
Private Deployment Options✓

Try it yourself

Test the voice with words it has Never Recorded

A useful voice clone should not only repeat the reference sentence. Give it something new to say.

01

Upload a voice sample

Use a short, clean recording of the speaker whose voice you are authorised to use.

Recommended length: 3-6 seconds

02

Reference transcript

Enter the exact words spoken in the reference audio.

Welcome to Shunya. This is my reference voice sample.

03

New text

Now enter something completely different.

Your appointment has been confirmed for tomorrow morning at ten.

04

Generate

Radiance uses the reference voice to synthesize the new sentence.

Clone this voice in the playground.

How voice cloning works

Reference audio in. New Speech Out.

Voice cloning uses characteristics from a reference recording to guide speech synthesis.

01

Record

Provide a clean sample of the speaker.

02

Transcribe

Provide the exact transcript of what was spoken in that reference clip.

03

Clone

The reference audio establishes the voice characteristics used during synthesis.

04

Enter new text

Write a sentence that was never recorded by the original speaker.

05

Generate speech

Radiance generates the new sentence using the reference voice.

  1. 01

    Reference audio

    The speaker sample

  2. 02

    Voice characteristics

    Timbre · Prosody · Accent

  3. 03

    New text

    Words the speaker never recorded

  4. 04

    Generated speech

    The new sentence, in the reference voice

The words can change. The voice identity remains connected to the reference.

The reference sample matters

Better input creates a Better Clone

Voice cloning quality depends heavily on the audio used as the reference. A noisy or distorted recording gives the model less reliable information about the speaker.

Clean speech

Use a recording where the speaker can be heard clearly.

Minimal background noise

Avoid traffic, music, other speakers, and environmental noise.

Consistent volume

The speaker should remain at a stable distance from the microphone.

Limited reverb

Avoid large echoing rooms where possible.

Clear pronunciation

The speaker should talk naturally but clearly.

Good audio quality

Use a sample rate of at least 16 kHz where possible.

3-6 seconds

Recommended length

Long recordings are not automatically better. A short, clean, representative voice sample is more useful than a longer recording filled with noise or inconsistent speech.

Reference text

Tell the model exactly what the Speaker Said

Voice cloning uses both the audio containing the reference speaker and the exact transcript of the reference recording. These two pieces of information work together.

reference_wav

The audio containing the reference speaker.

Hello, welcome to our customer support service.

reference_text

The exact transcript of the reference recording.

Hello, welcome to our customer support service.

New text

Your request has been completed successfully.

A sentence the reference recording never included.

Generated voice

Spoken with the reference speaker.

The new sentence is synthesized using characteristics from the reference speaker.

Match the transcript.The reference_text should accurately represent what is spoken in the reference recording. Incorrect reference text can reduce the quality of the resulting voice clone.

No fixed script

The clone can say New Things

The value of voice cloning is not repeating the reference recording. It is generating speech that did not exist before.

ReferenceWelcome to our service.

New sentence 01

Your payment has been processed successfully.

New sentence 02

Your appointment is scheduled for Monday at four PM.

New sentence 03

Your order has left our warehouse and should arrive tomorrow.

New sentence 04

Is there anything else I can help you with today?

One referenceMany new sentences

This makes voice cloning useful for applications where the content changes dynamically.

One voice, more languages

Voice identity shouldn't disappear whenthe Language Changes

A voice may begin in English. The application may later need Hindi, Tamil, Bengali, Telugu, or anothersupported language. Radiance supports cross-lingual synthesis for cloned voices.

One speaker

Reference voice

English

Your appointment is confirmed.

Hindi

आपकी अपॉइंटमेंट कन्फर्म हो गई है।

Tamil

உங்கள் சந்திப்பு உறுதிசெய்யப்பட்டுள்ளது.

Bengali

আপনার অ্যাপয়েন্টমেন্ট নিশ্চিত করা হয়েছে।

Same voice identity.The spoken language changes. The voice remains connected to the same reference speaker.

Cross-lingual voice cloning: This can help multilingual products maintain a more consistent voice experience instead of assigning a completely different speaker to every language.

Expressive cloned voices

Keep the voice. Change the Delivery.

Voice identity and speaking style are different controls. The same cloned voice may need to sound conversational in one interaction

and more narrative in another. Radiance provides multiple expression styles through the same speech stack.

Conversational

For voice agents, customer interactions, and everyday dialogue.

Narrative

For stories, learning material, articles, and long-form speech.

News

For structured informational delivery.

Happy

For positive confirmations and friendly interactions.

Enthusiastic

For higher-energy speech.

Sad

For more subdued delivery.

11

Expression styles

Same voiceConversationalNarrativeNewsHappyEnthusiasticDifferent delivery

The application can vary how speech is spoken without changing who is speaking.

Real-time voice cloning workflows

A cloned voice that can respond Dynamically

Voice cloning becomes particularly useful when combined with applications that generate new text continuously. A voice agent is one example.

UserCan you check the status of my order?HorizonConverts the customer's speech into text.LLM / AgentGenerates a new response. Your order was shipped this morning and should arrive tomorrow.RadianceUses the configured or cloned voice.User hearsThe dynamically generated response in the selected voice identity.

Listen → Think → Speak: A cloned voice can become the speaking layer of a conversational AI application.

Built-in voice vs. cloned voice

Choose the approach your product Actually Needs

Not every application needs voice cloning.

Use a built-in voice when

Launch with the existing catalogue

  • You need to launch quickly.
  • You do not need a unique voice identity.
  • A standard male or female speaker fits the product.
  • You want access to Shunya's existing voice catalogue.

Use a cloned voice when

The speaker identity is part of the product

  • You have permission to use a specific speaker's voice.
  • The voice itself is part of the product experience.
  • You need new speech without repeatedly recording the speaker.
  • You want a more consistent voice identity across dynamic content.
  • You need that voice to expand across supported languages.
RequirementBuilt-in voiceCloned voice
Reference Audio Needed-✓
Ready Immediately✓-
Unique Speaker Identity-✓
Dynamic New Text✓✓
Expression Control✓✓
Cross-lingual Speech✓✓
Streaming Workflows✓✓

Different entry point.Same Radiance speech infrastructure.

Responsible voice cloning

A voice is part of a Person's Identity

Voice cloning should only be used when you have the appropriate rights and permission to use the reference voice.

01

Confirm voice-owner consent

The person whose voice is being cloned should understand and authorise its synthetic use.

02

Confirm recording rights

Make sure you have the rights necessary to use the original reference recording.

03

Define the permitted use

A voice approved for one project should not automatically be assumed to be approved for every other use.

04

Protect reference audio

Treat reference recordings as identity-related source material and manage access appropriately.

Control the voice: Voice cloning is a production capability. Consent and authorised use should be part of the workflow, not an afterthought.

Real-world use cases

Voice cloning where Consistency Matters

AI Voice Agents

Give conversational AI a specific authorised voice identity while responses change dynamically.

Customer Experiences

Maintain a recognisable voice across automated conversations, support journeys, and notifications.

Multilingual Applications

Use one authorised voice across supported languages instead of changing speaker identity between markets.

Media & Dubbing

Generate authorised speech for translated or adapted content while maintaining a consistent speaker identity.

E-learning

Create new educational speech without recording every lesson sentence individually.

Accessibility

Build personalised speech experiences using an authorised reference voice.

Interactive Characters

Maintain a consistent synthetic character voice across changing dialogue.

Enterprise Assistants

Give internal or customer-facing assistants a controlled organisational voice identity.

Announcements

Generate changing announcements through the same approved voice.

Dynamic Content

Create audio from newly generated text without scheduling another recording session.

Developers

Clone a voice inside the TTS Request

Radiance does not require a separate voice-registration step for the inline cloning workflow. The reference can be supplied directly during synthesis.

POST /v1/audio/speech

{
  "model": "radiance-indic",
  "input": "Your request has been completed successfully.",
  "voice": "Varun",
  "reference_wav": "<base64-encoded audio>",
  "reference_text": "Hello, welcome to our customer support service."
}

Input

New text to synthesize

The sentence the cloned voice should speak.

reference_wav

Reference audio

Base64-encoded WAV, FLAC, or OGG clip of the reference speaker.

reference_text

Reference transcript

The exact text spoken in the reference audio.

model

Model

Generated speech using characteristics from the reference voice.

Reference WAVReference textNew textPOST /v1/audio/speechCloned voice audio

Important.For the inline workflow, the reference is used for the request. A separate permanent voice-registration step is not required.

Shunya-specific parameters: Voice cloning uses reference_wav and reference_text. The voice field is still required, and the reference audio overrides that preset speaker. These fields are available through the Shunya API.

Explore Voice Cloning API

Control the generated voice

Clone the identity. Control the Speech.

After the voice is established, the application still needs control over the generated output.

Voice identity

Reference sample

The authorised speaker clip that establishes who the generated voice sounds like.

Speech control

Language, expression, speed, format

Choose how the cloned voice speaks and which audio encoding the application receives.

Final audio

Generated output

Speech synthesized from new text while staying connected to the reference voice.

Text

Choose what the voice says.

Language

Choose the supported language being spoken.

Expression

Choose how the sentence is delivered.

Speed

Control speaking speed.

FormatCommon use
MP3Web playback and general applications
WAVHigh-quality audio workflows
PCMLow-latency and real-time pipelines
OGG OpusEfficient streaming and web audio
FLACLossless audio
μ-lawTelephony and IVR
A-lawTelephony and SIP systems

The voice clone becomes one component of a configurable speech-generation workflow.

Multilingual voice experiences

One authorised voice across more of the Conversation

Voice cloning can be combined with the rest of the Shunya speech stack.

Speech stack

Speech recognitionHorizon - understand what the user says.Language processingLLM / Parallax - generate or translate the response.Voice generationRadiance - speak using the authorised cloned voice.

Example

Customer speaks HindiSTT transcribes HindiApplication generates a responseCloned voice speaks the response

One voice layer.The same voice workflow can become part of multilingual assistants, translated media, customer support, and other speech applications.

Built for production

Beyond the Voice-Cloning Demo

A convincing five-second demo is not the same as operating a voice product. Production applications need control overreference quality, generated text, language, expressions, audio formats, APIs, deployment, and authorised use.

RequirementRadiance
Voice Cloning✓
Short Reference Audio3-6 sec
Reference Transcript✓
Dynamic New Speech✓
Cross-lingual Synthesis✓
Expression Control✓
Speed Control✓
Batch Synthesis✓
Streaming TTS✓
Multiple Audio Formats✓
Telephony Formats✓
API Integration✓
Voice Agent Integration✓
Private Deployment Options✓

One voice

Reference

ReferenceMany experiences
Voice agentMediaEducationCustomer supportMultilingual application

Same voice identity

Voice cloning becomes infrastructure for applications that need a consistent authorised speaker rather than a one-time audio effect.

Questions

Frequently Asked Questions

What is voice cloning?

Voice cloning uses a reference recording of a speaker to generate new synthetic speech that follows characteristics of that reference voice.

Does Shunya support voice cloning?

Yes. Radiance supports voice cloning using reference audio and its transcript.

How much audio do I need?

Shunya's current documentation recommends approximately 3-6 seconds of clean reference audio.

What kind of reference recording works best?

Use clear speech with minimal background noise, limited reverb, consistent volume, and a sample rate of at least 16 kHz where possible.

Do I need the transcript of the voice sample?

Yes. The inline cloning workflow uses both reference_wav and reference_text. The reference text should accurately match what the speaker says in the audio.

Can the cloned voice say words that were not in the sample?

Yes. The reference establishes the voice characteristics, while the new input text determines what speech is generated.

Does Shunya permanently store the reference voice?

In the inline voice-cloning workflow, there is no separate registration step. The reference audio and transcript are supplied as part of the synthesis request.

Can a cloned voice speak another language?

Yes. Radiance supports cross-lingual synthesis for cloned voices across supported languages.

Can I change the expression of a cloned voice?

The voice cloning workflow sits inside Radiance, which supports multiple expression styles for speech generation.

Can I change speaking speed?

Yes. Radiance provides configurable speech-speed control.

Can I use a cloned voice in a voice agent?

Yes. A cloned voice can be used as the TTS layer after an LLM or conversational agent generates its response.

Can I use voice cloning for multilingual applications?

Yes. Cross-lingual synthesis allows one voice identity to be used across supported languages.

Which audio formats can the generated voice use?

Radiance supports MP3, WAV, PCM, OGG Opus, FLAC, μ-law, and A-law output.

Is voice cloning available through the OpenAI-compatible SDK?

The standard OpenAI speech SDK interface does not expose Shunya-specific voice-cloning parameters such as reference_wav and reference_text. Use the Shunya SDK or REST API when those capabilities are required.

Do I need permission to clone someone's voice?

You should only clone and synthesize a person's voice when you have the appropriate rights to the reference recording and the voice owner's consent for the intended use.

Are cloned voices available on every Shunya plan?

Shunya's current pricing page lists custom and cloned voices under Enterprise availability. Check the current plan or speak with Shunya before planning a production deployment around voice cloning.

Get Started

Clone a voice. Give it New Words

Don't test voice cloning with our speaker. Test an authorised voice your application actually needs. Provide a cleanreference clip, enter its transcript, then give the voice something completely new to say. Try another sentence.Change the language. Change the expression. Hear whether the voice identity stays consistent.