Яндекс.Метрика

Polyglot Voice

Vous lisez un guide Polyglot Voice — notre plateforme pour transformer la parole en texte et la traduire.

Pour transcrire ou traduire un audio ou vidéo en texte, connectez-vous ou créez un compte gratuit (cela ne prend qu'une minute). Le studio prend en charge 98 langues ; vous pouvez changer de direction et de modèles à tout moment après le téléchargement.

Speech-to-Text API

Use the API when transcription and translation should run inside your product, bot, internal tool or automation. Create tasks from files or links, track status, receive webhooks and fetch text results.

Upload audio/video or create tasks from supported URLs

Receive completion webhooks and fetch structured results

Connect transcription to translation, subtitles and automation

Try the workflow before paying

From recording to text, translation or voiceover in a few steps

Give visitors a clear path: upload a file or record speech, get text, then translate, dub or export the result. The free start is enough to understand the quality.

AI speechTranslationSubtitles

Quick price estimate

10 min
10 000

Transcription from: 490

Voiceover approx.: 10

This is an estimate: final cost depends on quality mode, pack and tariff.

How it works

  1. Upload audio/video, paste a link or record speech in the browser.
  2. Choose language, quality and the target result: text, translation, subtitles or voiceover.
  3. Get the result in the studio and add minutes, TTS characters or API access when needed.

Example result

Before

Before: a lecture recording, interview or video in another language.

After

After: structured text, translation, subtitles and a base for voiceover or clips.

Why users can trust it

  • Source files are removed from disk about 1 hour after processing.
  • Start for free and pay only for the minutes, attempts or TTS characters you need.
  • Use the web studio for one-off jobs or API access for automation.

What plans include

  • Transcription minutes/attempts
  • TTS characters for voiceover
  • Task history, export, API keys and webhooks

Honest limitations

  • Demo is limited to 60 seconds
  • Quality depends on noise and speech clarity
  • Long files and heavier models require a plan

Comment ça marche

  1. Téléchargez le fichier audio ou vidéo source correspondant à votre flux de travail.
  2. Choisissez la tâche, la langue ou la direction de traduction correspondant à votre objectif.
  3. Vérifiez le résultat, exportez la sortie et continuez avec les sous-titres, les clips ou l'automatisation API.

Formats et entrées pris en charge

REST APImultipart uploadURL taskswebhooks

Idéal pour

  • voice message bots
  • call processing
  • media archives
  • internal automation

FAQ

How do I start using the speech-to-text API?

Create an API key in the dashboard, send an upload request and fetch the result when the task is completed.

Does the API support webhooks?

Yes. You can configure a completion webhook to receive events when transcription tasks finish.

Can I transcribe video through the API?

Yes. The API accepts supported audio/video files and URL-based tasks depending on your plan and limits.

Guides associés

Pages associées