huggingface/speech-to-speech

Skip(New discovery — enrichment pending)

Build local voice agents with open-source models

Why this repo matters

Provides end-to-end speech-to-speech translation pipeline preserving original speaker voice characteristics and prosody.
Integrates seamlessly with Hugging Face ecosystem for rapid prototyping of multilingual voice interaction systems.
Offers open weights and code enabling local deployment without proprietary API dependencies or latency issues.

Key Metrics

Stars: 13k
Forks: 1.6k
Open Issues: 101
Stars (7d delta): 0
Stars (30d delta): 0

Scores & Metadata

Trend Velocity Score: 0.00
Opportunity Score: 10.00
Confidence Score: 100.00
Language: Python
License: Apache-2.0
Topics:
🏷️ ai 🏷️ assistant 🏷️ language-model 🏷️ machine-learning 🏷️ python 🏷️ speech 🏷️ speech-synthesis 🏷️ speech-to-text 🏷️ speech-translation
First Seen: 3 months ago

Star History

13.0k stars+6.9k over 15 snapshots7/17/2026 – 9/5/2026

Links & Sources

Related Repositories

codecrafters-io/build-your-own-x

547.5kMarkdown

freeCodeCamp/freeCodeCamp

freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.

456.9kTypeScript

affaan-m/ECC

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

269kJavaScript

multica-ai/andrej-karpathy-skills

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

217.5k