v1.2.1

What's Changed

• Split verbatim into verbatim_cli and verbatim_serve by @gaspardpetit in #232
• Reusable CLI arguments by @gaspardpetit in #233
• Added batch interface by @gaspardpetit in #234
• Support config file by @gaspardpetit in #236
• stereo diarizer is now energy / channel diarizer by @gaspardpetit in #237
• defer downmixing until pyannote strategy by @gaspardpetit in #238
• Added support for complex diarization policies by @gaspardpetit in #239
• Feat/speaker by @gaspardpetit in #240
• Lazy load heavy dependencies by @gaspardpetit in #241
• Transcribe according to diarization groups by @gaspardpetit in #242
• Added word-level timestamps and language detection in mlx-whisper tra… by @gaspardpetit in #243

Full Changelog: v1.2.0...v1.2.1