May 09, 2025
  • About
  • Contact Us
  • Greek (Greece)
  • English (UK)
ID-on't renounce my freedomID-on't renounce my freedom
  • Articles
    • Privacy
      • General Privacy Issues
      • Social Μedia
      • Data Mining
      • Data Brokers
      • Data Leaks - Hackers
      • Surveilance
      • Face Recognition
    • e- Government
      • General e-Government Issues
      • Citizen Card
      • Social Credit
    • Artificial Intelligence
      • General Artificial Intelligence Issues
      • Man Machine Interface
      • Robots
      • Superhumanism and Rehumanism
    • Economy
      • Cashless Society
    • Legal Issues
      • Legal News
    • Technologies
      • Bar Code
      • RFID
      • NFC
      • Electronic Implants
      • Cryptography
  • Blog
  • Videos
  • Books
  • Laws
  • You are here:  
  • Home
  • Articles
  • Artificial Intelligence
  • Google’s AI can now translate your speech while keeping your voice

Google’s AI can now translate your speech while keeping your voice

Emilio Marenatti
Artificial Intelligence
24 December 2019
Hits: 2251
  • Print
  • Email

google tranlation 01

Listen to this Spanish audio clip.

Download the audio

This is how its English translation might sound when put through a traditional automated translation system.

Download the audio

Now this is how it sounds when put through Google’s new automated translation system.

Download the audio

The results aren’t perfect, but you can sort of hear how Google’s translator was able to retain the voice and tone of the original speaker. It can do this because it converts audio input directly to audio output without any intermediary steps. In contrast, traditional translational systems convert audio into text, translate the text, and then resynthesize the audio, losing the characteristics of the original voice along the way.The new system, dubbed the Translatotron, has three components, all of which look at the speaker’s audio spectrogram—a visual snapshot of the frequencies used when the sound is playing, often called a voiceprint.  The first component uses a neural network trained to map the audio spectrogram in the input language to the audio spectrogram in the output language. The second converts the spectrogram into an audio wave that can be played. The third component can then layer the original speaker’s vocal characteristics back into the final audio output.

Not only does this approach produce more nuanced translations by retaining important nonverbal cues, but in theory it should also minimize translation error, because it reduces the task to fewer steps.

Translatotron is currently a proof of concept. During testing, the researchers trialed the system only with Spanish-to-English translation, which already took a lot of carefully curated training data. But audio outputs like the clip above demonstrate the potential for a commercial system later down the line. You can listen to more of them here.

 

Source: google-research.github.io

 

Tags: Man Machine Interface
  • Prev
  • Next

Follow Us

  • RSS Subscribe us on News
  • Facebook Like us on Facebook
  • Twitter Follow us on Twitter
  • Youtube Subscribe on Youtube

Popular Articles

Error: No articles to display

footer-logo.png

The 'ID-on't renounce my freedom' website contains articles and news related to the growing threat to our personal freedom and privacy.

  info@id-ont.org

© 2019 ID-on't renounce my freedom | Designed by Privacy Team
  • Privacy
  • Site Terms
  • Contact Us
  • Home
  • Articles
    • Privacy
    • e- Government
    • Artificial Intelligence
    • Economy
    • Legal Issues
    • Technologies
  • Blog
  • Videos
  • Books
  • Laws