diff --git a/.fern/metadata.json b/.fern/metadata.json index 0f155d5..f4f7a0d 100644 --- a/.fern/metadata.json +++ b/.fern/metadata.json @@ -14,5 +14,5 @@ }, "exclude_types_from_init_exports": true }, - "originGitCommit": "37ba641a90aa4496eb97146f05555706aaf92670" + "originGitCommit": "3fdb45ab45802742325afca544fdc22f5ae58db8" } \ No newline at end of file diff --git a/.fern/replay.lock b/.fern/replay.lock index 9c78b96..a1ce605 100644 --- a/.fern/replay.lock +++ b/.fern/replay.lock @@ -60,14 +60,20 @@ generations: cli_version: unknown generator_versions: fernapi/fern-python-sdk: 4.37.0 -current_generation: 48b0250bddbc7acc02cd9da9fa06c1fefe1dd0f7 + - commit_sha: 3885ea61224cbf0671c305350b8d1cddef509443 + tree_hash: 2b9a17c0608360e1eb6044e97e414ae3d936d857 + timestamp: 2026-09-09T15:14:03.378Z + cli_version: unknown + generator_versions: + fernapi/fern-python-sdk: 4.37.0 +current_generation: 3885ea61224cbf0671c305350b8d1cddef509443 patches: - id: patch-13d8e068 content_hash: sha256:97d879ab169016b9abaf1866427a8ebc45af608f69e063a07e5126c9613c329b original_commit: 13d8e0683ffa08bcfc9a381eb9604adf5abdbe0b original_message: "chore: ignore local venv and planning docs" original_author: plutoless - base_generation: 48b0250bddbc7acc02cd9da9fa06c1fefe1dd0f7 + base_generation: 3885ea61224cbf0671c305350b8d1cddef509443 files: - .gitignore patch_content: | @@ -96,7 +102,7 @@ patches: original_commit: a065088b4c45e2ca7e1bb91b1a899b8444800121 original_message: "fix(vendors): send Speechmatics STT key instead of api_key" original_author: digitallysavvy - base_generation: 48b0250bddbc7acc02cd9da9fa06c1fefe1dd0f7 + base_generation: 3885ea61224cbf0671c305350b8d1cddef509443 files: - src/agora_agent/types/speechmatics_asr_params.py patch_content: | diff --git a/src/agora_agent/types/gemini_asr_params.py b/src/agora_agent/types/gemini_asr_params.py index 0e9714a..ef064ef 100644 --- a/src/agora_agent/types/gemini_asr_params.py +++ b/src/agora_agent/types/gemini_asr_params.py @@ -5,6 +5,7 @@ import pydantic from ..core.pydantic_utilities import IS_PYDANTIC_V2 from ..core.unchecked_base_model import UncheckedBaseModel +from .gemini_asr_params_mode import GeminiAsrParamsMode class GeminiAsrParams(UncheckedBaseModel): @@ -32,9 +33,29 @@ class GeminiAsrParams(UncheckedBaseModel): The language code for speech recognition. This takes precedence over the top-level `asr.language` value. """ + language_hints: typing.Optional[typing.List[str]] = pydantic.Field(default=None) + """ + Candidate language codes for transcription. When non-empty, these take precedence over language. + """ + + custom_vocabulary: typing.Optional[typing.List[str]] = pydantic.Field(default=None) + """ + Words and phrases used to bias transcription. A non-empty custom vocabulary cannot be combined with word_timestamp. + """ + + mode: typing.Optional[GeminiAsrParamsMode] = pydantic.Field(default=None) + """ + Transcription output mode. SMART removes disfluencies and applies formatting; VERBATIM preserves literal speech. When omitted, the service defaults to VERBATIM. SMART cannot be combined with word_timestamp or diarization. + """ + word_timestamp: typing.Optional[bool] = pydantic.Field(default=None) """ - Whether to include word-level timestamps in the transcription results. + Whether to include word-level timestamps in the transcription results. Cannot be enabled when mode is SMART or custom_vocabulary is non-empty. + """ + + diarization: typing.Optional[bool] = pydantic.Field(default=None) + """ + Whether to include speaker labels in the transcription results. Cannot be enabled when mode is SMART. """ if IS_PYDANTIC_V2: diff --git a/src/agora_agent/types/gemini_asr_params_mode.py b/src/agora_agent/types/gemini_asr_params_mode.py new file mode 100644 index 0000000..75bbc7a --- /dev/null +++ b/src/agora_agent/types/gemini_asr_params_mode.py @@ -0,0 +1,5 @@ +# This file was auto-generated by Fern from our API Definition. + +import typing + +GeminiAsrParamsMode = typing.Union[typing.Literal["SMART", "VERBATIM"], typing.Any]