v2.24.0

What's Changed

Added

• Deepgram speech-to-text provider — selectable via AIFactory.create_speech_to_text("deepgram", "nova-3"), complementing the existing Deepgram TTS. Posts to /v1/listen, auth via DEEPGRAM_API_KEY, normalizes to TranscriptionResponse with text + segments (Deepgram extras under each segment's metadata; None when no timing returned). (#202)

Fixed

• OpenAI STT gpt-4o-transcribe compatibility — no longer hardcodes response_format="verbose_json"; sends verbose_json for the Whisper family and json for the gpt-4o-transcribe / gpt-4o-mini-transcribe family, which previously failed. Segments/duration stay None when those models omit them. (#204)
• mypy --strict re-exports — package-level __all__ is now a static literal, so downstream packages with mypy --strict can from esperanto import AIFactory without "does not explicitly export attribute" errors. Runtime behavior unchanged. (#190)

Full Changelog: v2.23.0...v2.24.0