Multilingual Speech Sample from Global Contributor Network

A sample from the Silencio corpus, which contains over 100,000 hours of speech data. The full dataset is collected from a community of over 2 million contributors across more than 180 countries and 100 languages.

Use Cases

• Train acoustic models on real-world audio data from 100+ languages for multilingual speech recognition.
• Develop voice biometrics systems using speech samples collected from diverse accents and dialects across 180+ countries.
• Build models for natural language processing tasks using audio data from a large-scale, opt-in contributor network.

Strengths

• The full corpus contains over 100,000 hours of speech data.
• Data is collected from a community of over 2 million contributors.
• The dataset covers more than 180 countries and 100 languages.
• Data is collected 'in the wild', representing real-world acoustic conditions.

Limitations

• This is a sample dataset; the full corpus is not directly available here.
• Specific details on audio quality, file formats, and licensing are unknown.
• The sample size, duration, and specific language/geographic breakdowns are not provided.

Provenance

Source

SilencioNetwork

Collection Method

Collected 'in the wild' from a massive, opt-in community of contributors.

Freshness

The dataset page was last updated on 2026-02-27.

Geography

Global coverage across 180+ countries.