Real-Time Speech to Speech Translator API
Palabra.ai gives you an out-of-the-box two-way speech-to-speech translation API. Powered by our own models, it delivers human-level accuracy with <1s latency.
Language auto-detection
Tone and context conveyed
Voice cloning out of the box
Speaker diarization (coming soon)
Ultra-low latency
Custom glossaries
Enterprise-grade security
Feedback on our real-time
translation services
“At EventLabs, we rely on Palabra for real-time translation during conferences and live events. Among all the solutions we’ve tested, Palabra stands out with the highest translation quality and the lowest latency by a significant margin. The platform’s speaker autodetection, differentiating between male and female voices and adapting translations in real time, has noticeably improved the listener experience. For us, Palabra is setting the benchmark for event translation technology.”
“We built our product entirely on the Palabra API and it’s been an incredible foundation for what we do at Talo. The API’s natural language processing capabilities are reliable and accurate, which allowed us to bring real-time translations and captions to our users without starting from scratch.”
Designed for what you build
Impress your speakers and guests with Live Translation powered by Palabra’s very own language models, offering state-of-the-art accuracy and small latency
custom glossaries.
What teams build with the Palabra.ai real-time voice translation API
Integrate real-time speech translation with just a few lines of code
from palabra_ai import (PalabraAI, Config,
SourceLang, TargetLang, EN, ES, DeviceManager)
palabra = PalabraAI('<API_CLIENT_ID>', '<API_CLIENT_SECRET>')
dm = DeviceManager()
mic, speaker = dm.select_devices_interactive()
cfg = Config(SourceLang(EN, mic), [TargetLang(ES, speaker)])
palabra.run(cfg)
import { PalabraClient, getLocalAudioTrack } from '@palabra-ai/translator';
const client = new PalabraClient({
auth: {
clientId: 'YOUR_CLIENT_ID',
clientSecret: 'YOUR_CLIENT_SECRET',
},
translateFrom: 'en', // Source language code
translateTo: 'es', // Target language code handleOriginalTrack: getLocalAudioTrack, // Func returning a MediaStreamTrack
});
import ai.palabra.*;
import ai.palabra.adapter.*;
public class TranslationExample {
public static void main(String[] args) {
// Initialize client with credentials
String clientId = System.getenv("PALABRA_CLIENT_ID");
String clientSecret = System.getenv("PALABRA_CLIENT_SECRET");
PalabraAI client = new PalabraAI(clientId, clientSecret);
// Configure translation
Config config = Config.builder()
.sourceLang(Language.EN_US)
.targetLang(Language.ES_MX)
.reader(new FileReader("input.wav"))
.writer(new FileWriter("output.wav"))
.build();
// Run translation
client.run(config);
}
}
Try it in your code in 4 steps
with Palabra voice translation API
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
// clientId: "<API_CLIENT_ID>",
// clientSecret: "<API_CLIENT_SECRET>",
// originalTrack: getMicAudioTrack(),
// translateFrom: "en",
// translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
// originalTrack: getMicAudioTrack(),
// translateFrom: "en",
// translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
originalTrack: getMicAudioTrack(),
translateFrom: "en",
translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
originalTrack: getMicAudioTrack(),
translateFrom: "en",
translateTo: "fr"
});
document.getElementById('start')
.addEventListener('click', () => {
client.startTranslation();
client.playTranslationTrack();
});
document.getElementById('stop')
.addEventListener('click', () => {
client.stopTranslation();
});
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
// clientId: "<API_CLIENT_ID>",
// clientSecret: "<API_CLIENT_SECRET>",
// originalTrack: getMicAudioTrack(),
// translateFrom: "en",
// translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
// originalTrack: getMicAudioTrack(),
// translateFrom: "en",
// translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
originalTrack: getMicAudioTrack(),
translateFrom: "en",
translateTo: "fr"
});
// document.getElementById('start')
// .addEventListener('click', () => {
// client.startTranslation();
// client.playTranslationTrack();
// });
// document.getElementById('stop')
// .addEventListener('click', () => {
// client.stopTranslation();
// });
HTML
1
2
3
4
5
6
7
8
<div class="app">
<div class="transcription" />
<div class="controls">
<button id="start">Start</button>
<button id="start">Stop</button>
</div>
<div class="translations" />
</div>
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
import { PalabraClient, getMicAudioTrack }
from "palabra";
const client = new PalabraClient({
clientId: "<API_CLIENT_ID>",
clientSecret: "<API_CLIENT_SECRET>",
originalTrack: getMicAudioTrack(),
translateFrom: "en",
translateTo: "fr"
});
document.getElementById('start')
.addEventListener('click', () => {
client.startTranslation();
client.playTranslationTrack();
});
document.getElementById('stop')
.addEventListener('click', () => {
client.stopTranslation();
});
at [email protected] or book a demo call.
Choose your plan
Yearly (2 months for free)
Contact us
Answers you might need
What industries can benefit most from a real-time voice translation API?
Industries that benefit most include customer service, enterprise software and collaboration, media and entertainment, and consumer apps.
How do I integrate the API into my existing application?
You can integrate Palabra API into your application by using our SDKs or connecting directly via WebRTC (for browsers) or WebSockets (for servers).
Which programming languages and SDKs are supported?
Palabra provides SDKs and client libraries for Python and JavaScript. For other languages, Palabra integrates through WebRTC (frontend) and WebSockets (backend).
Is HTML & XML handling supported for translation?
As a real-time speech-to-speech translation solution, Palabra supports audio input only.
Can I customize translations with a glossary?
Yes. Glossaries let you define how Palabra translates specific terms. Once enabled, your glossary applies across all Palabra applications and sessions.
How is data secured during and after translation?
All conversations are encrypted in transit and processed entirely in memory. Palabra does not store voice data on its servers ー once audio is translated, it is deleted.
Does the API store or log voice data?
No. By default, Palabra does not store or log user data, nor is user data used to train models.
Can the API be deployed in a private cloud or on-premises environment?
Yes. The Palabra API can run in a private cloud or on-premises, fully under your own security and compliance controls.
What audio does Palabra.ai support?
Palabra.ai WebSocket integration supports these input audio formats: Opus, PCM_S16LE, and WAV. For output, it supports PCM_S16LE and ZLIB_PCM_S16LE.
What is the maximum audio length supported per request?
Palabra processes real-time audio streams, which by default can run indefinitely.
How does the API maintain accuracy in noisy environments?
Palabra includes integrated noise suppression, so speech remains accurate even in noisy conditions. No additional preprocessing is required.
.svg)
