Inworld open-sources voice isolation model Voice Isolator
What happened
Inworld released its first open-source model, Voice Isolator, which removes background voices, music, and transient noise while preserving the speaker's timbre, breathing, and volume. The model operates at 48 kHz and offers two checkpoints: an offline version aimed at existing recordings and best quality, and a real-time version that streams with 20 ms lookahead for live scenarios such as streaming. Model weights are downloaded from Hugging Face on first use and can run on CUDA, Apple silicon, and CPU. Inworld says VIBE is currently a private benchmark.
AI-written from the coverage · updated 3 d ago
Coverage timeline
Follow the coverage to see the story from different sides.
- Inworld AIPickInworld open-sources Voice Isolator and SAVA speech models and publishes TTS-2 technical report
Inworld open-sourced two speech models, Voice Isolator and SAVA (Semantic-Acoustic Voice Activity), and published the technical report behind Realtime TTS-2.
- Inworld AI BlogPickInworld open-sources voice isolation model Voice Isolator with offline and real-time versions
Inworld released Voice Isolator, its first open-source model, which removes background voices, music, and transient noise while preserving the speaker's timbre, breathing, and volume. It works at 48 kHz and comes in two checkpoints: offline and real-time streaming with 20 ms lookahead.
Heat trend for this story
- Current (comparable)
- 3
- Peak (comparable)
- 6Nov 19 02:00
- 24-hour change
- -49%
The trend only compares participants observed continuously and completely, so it may cover fewer than the current heat count. Hover over or tap the chart to see each hour's heat; with a keyboard, use the left and right arrow keys.