v0.10.2

What's Changed

• Fix device selection by @twitchard in #330
• 🌿 Fern Regeneration -- August 5, 2025 by @fern-api[bot] in #337
  • Language Model Expansion: Added support for new language models across multiple providers, including gpt-oss-120b, gpt-4.1, qwen-3-235b-a22b, gemini-2.5-flash, moonshotai/kimi-k2-instruct
  • Enhanced Chat Events: Expanded ReturnChatEventType with 5 additional event types.

Full Changelog: v0.10.1...v0.10.2

v0.10.1

What's Changed

• 🌿 Fern Regeneration -- July 17, 2025 by @fern-api[bot] in #326
  • Fix: Added AssistantProsody to JSONMessage union.

Full Changelog: v0.10.0...v0.10.1

v0.10.0

What's Changed

• Release EVI 3.
  • break: EVI2 hume.custom_voices was removed - use EVI3 and hume.tts.voices instead.
  • Add support forAssistantProsody messages. For EVI3 chats, expression measures will no longer be contained directly on assistant messages, subscribe to the assistant_prosody message type instead.

Full Changelog: v0.9.1...v0.10.0

v0.9.1

What's Changed

• (fix) Remove dependencies that aren't used anymore by @twitchard in #317 and #319
  • More completely remove the optional dependencies that were used in the extras removed in https://github.com/HumeAI/hume-python-sdk/releases/tag/v0.9.0

Full Changelog: v0.9.0...v0.9.1

v0.9.0

What's Changed

• Make audio playback more reliable and easier to install @twitchard in #303
  • Removed dependency on pydub
  • Add support for streaming audio playback.
  • Improve compatibility across more systems.
• Remove legacy extra by @twitchard in #311
  • Users reliant on the legacy package extra should remain on v0.8.6 or follow the instructions in our migration guide to use the new interfaces.

Full Changelog: v0.8.6...v0.9.0

v0.8.6

What's Changed

• 🌿 Fern Regeneration -- June 13, 2025 by @fern-api in #309
  • This release contains changes from upgrading our Fern python generator from 4.2.8 to 4.21.5. See the fern changelog. Some highlights:
    • Retries now default to 2 instead of 0.
    • Error reporting and Pydantic types are improved.
    • Paginating without an explicit page number now works correctly!
  • API addition: support defining nudges in your Evi config, to help Evi naturally fill the silence during inactivity.
  • API addition: TTS responses now contain transcribed_text on Snippet when instant_mode is not enabled.
• Respect base_url by @twitchard in #306

Full Changelog: v0.8.5...v0.8.6

v0.8.5

What's Changed

• 🌿 Fern Regeneration -- June 9, 2025 by @fern-api in #301
  • (fix) fix version of dependency websockets to 13.1
  • (fix) More informative audio import errors by @twitchard in #302

Full Changelog: v0.8.4...v0.8.5

v0.8.4

What's Changed

• fix: fixed a bug where auto-pagination skipped from page 0 to page 2.
• fix: removed unreferenced FunctionCallResponseInput, PostedPromptSpec, TextInput and TtsInput types.
• fix: fixed incorrectly reported package version in request headers.

Full Changelog: v0.8.3...v0.8.4

v0.8.3

What's Changed

• Bump websockets dependency 12 -> 13.1 by @twitchard in #284

Full Changelog: v0.8.2...v0.8.3

v0.8.2

What's Changed

• 🌿 Fern Regeneration -- April 18, 2025 by @fern-api in #282
  • Add strip_headers to TTS streaming JSON apis, as a convenience when you want to play back audio as a single continued audio file vs a series of standalone audio files.

Full Changelog: v0.8.1...v0.8.2