NVIDIA Nemotron ASR Multilingual

Qwen Audio 3.0 TTS Flash

Fast multilingual text-to-speech with a broad set of voices and explicit language control.

ElevenLabs Scribe V2

ElevenLabs Scribe V2 transcription with improved accuracy, word-level timestamps, and speaker identification

ElevenLabs Scribe V1

ElevenLabs Scribe V1 transcription with word-level timestamps and speaker identification

ElevenLabs Turbo V2.5

High quality with lowest latency, ideal for real-time applications. Supports 32 languages while maintaining natural voice quality.

ElevenLabs v3

High-quality text-to-speech with enhanced controls and natural voices.

Demucs Stem Separation

Separate up to 60 minutes / 700 MiB of stereo audio into vocals, drums, bass, guitar, piano, and other stems with Demucs.