lfx-nightly 0.2.0.dev25__py3-none-any.whl

lfx/components/assemblyai/assemblyai_lemur.py

@@ -0,0 +1,183 @@
+import assemblyai as aai
+
+from lfx.custom.custom_component.component import Component
+from lfx.io import DataInput, DropdownInput, FloatInput, IntInput, MultilineInput, Output, SecretStrInput
+from lfx.log.logger import logger
+from lfx.schema.data import Data
+
+
+class AssemblyAILeMUR(Component):
+    display_name = "AssemblyAI LeMUR"
+    description = "Apply Large Language Models to spoken data using the AssemblyAI LeMUR framework"
+    documentation = "https://www.assemblyai.com/docs/lemur"
+    icon = "AssemblyAI"
+
+    inputs = [
+        SecretStrInput(
+            name="api_key",
+            display_name="Assembly API Key",
+            info="Your AssemblyAI API key. You can get one from https://www.assemblyai.com/",
+            advanced=False,
+            required=True,
+        ),
+        DataInput(
+            name="transcription_result",
+            display_name="Transcription Result",
+            info="The transcription result from AssemblyAI",
+            required=True,
+        ),
+        MultilineInput(name="prompt", display_name="Input Prompt", info="The text to prompt the model", required=True),
+        DropdownInput(
+            name="final_model",
+            display_name="Final Model",
+            options=["claude3_5_sonnet", "claude3_opus", "claude3_haiku", "claude3_sonnet"],
+            value="claude3_5_sonnet",
+            info="The model that is used for the final prompt after compression is performed",
+            advanced=True,
+        ),
+        FloatInput(
+            name="temperature",
+            display_name="Temperature",
+            advanced=True,
+            value=0.0,
+            info="The temperature to use for the model",
+        ),
+        IntInput(
+            name="max_output_size",
+            display_name=" Max Output Size",
+            advanced=True,
+            value=2000,
+            info="Max output size in tokens, up to 4000",
+        ),
+        DropdownInput(
+            name="endpoint",
+            display_name="Endpoint",
+            options=["task", "summary", "question-answer"],
+            value="task",
+            info=(
+                "The LeMUR endpoint to use. For 'summary' and 'question-answer',"
+                " no prompt input is needed. See https://www.assemblyai.com/docs/api-reference/lemur/ for more info."
+            ),
+            advanced=True,
+        ),
+        MultilineInput(
+            name="questions",
+            display_name="Questions",
+            info="Comma-separated list of your questions. Only used if Endpoint is 'question-answer'",
+            advanced=True,
+        ),
+        MultilineInput(
+            name="transcript_ids",
+            display_name="Transcript IDs",
+            info=(
+                "Comma-separated list of transcript IDs. LeMUR can perform actions over multiple transcripts."
+                " If provided, the Transcription Result is ignored."
+            ),
+            advanced=True,
+        ),
+    ]
+
+    outputs = [
+        Output(display_name="LeMUR Response", name="lemur_response", method="run_lemur"),
+    ]
+
+    def run_lemur(self) -> Data:
+        """Use the LeMUR task endpoint to input the LLM prompt."""
+        aai.settings.api_key = self.api_key
+
+        if not self.transcription_result and not self.transcript_ids:
+            error = "Either a Transcription Result or Transcript IDs must be provided"
+            self.status = error
+            return Data(data={"error": error})
+        if self.transcription_result and self.transcription_result.data.get("error"):
+            # error message from the previous step
+            self.status = self.transcription_result.data["error"]
+            return self.transcription_result
+        if self.endpoint == "task" and not self.prompt:
+            self.status = "No prompt specified for the task endpoint"
+            return Data(data={"error": "No prompt specified"})
+        if self.endpoint == "question-answer" and not self.questions:
+            error = "No Questions were provided for the question-answer endpoint"
+            self.status = error
+            return Data(data={"error": error})
+
+        # Check for valid transcripts
+        transcript_ids = None
+        if self.transcription_result and "id" in self.transcription_result.data:
+            transcript_ids = [self.transcription_result.data["id"]]
+        elif self.transcript_ids:
+            transcript_ids = self.transcript_ids.split(",") or []
+            transcript_ids = [t.strip() for t in transcript_ids]
+
+        if not transcript_ids:
+            error = "Either a valid Transcription Result or valid Transcript IDs must be provided"
+            self.status = error
+            return Data(data={"error": error})
+
+        # Get TranscriptGroup and check if there is any error
+        transcript_group = aai.TranscriptGroup(transcript_ids=transcript_ids)
+        transcript_group, failures = transcript_group.wait_for_completion(return_failures=True)
+        if failures:
+            error = f"Getting transcriptions failed: {failures[0]}"
+            self.status = error
+            return Data(data={"error": error})
+
+        for t in transcript_group.transcripts:
+            if t.status == aai.TranscriptStatus.error:
+                self.status = t.error
+                return Data(data={"error": t.error})
+
+        # Perform LeMUR action
+        try:
+            response = self.perform_lemur_action(transcript_group, self.endpoint)
+        except Exception as e:  # noqa: BLE001
+            logger.debug("Error running LeMUR", exc_info=True)
+            error = f"An Error happened: {e}"
+            self.status = error
+            return Data(data={"error": error})
+
+        result = Data(data=response)
+        self.status = result
+        return result
+
+    def perform_lemur_action(self, transcript_group: aai.TranscriptGroup, endpoint: str) -> dict:
+        logger.info("Endpoint:", endpoint, type(endpoint))
+        if endpoint == "task":
+            result = transcript_group.lemur.task(
+                prompt=self.prompt,
+                final_model=self.get_final_model(self.final_model),
+                temperature=self.temperature,
+                max_output_size=self.max_output_size,
+            )
+        elif endpoint == "summary":
+            result = transcript_group.lemur.summarize(
+                final_model=self.get_final_model(self.final_model),
+                temperature=self.temperature,
+                max_output_size=self.max_output_size,
+            )
+        elif endpoint == "question-answer":
+            questions = self.questions.split(",")
+            questions = [aai.LemurQuestion(question=q) for q in questions]
+            result = transcript_group.lemur.question(
+                questions=questions,
+                final_model=self.get_final_model(self.final_model),
+                temperature=self.temperature,
+                max_output_size=self.max_output_size,
+            )
+        else:
+            msg = f"Endpoint not supported: {endpoint}"
+            raise ValueError(msg)
+
+        return result.dict()
+
+    def get_final_model(self, model_name: str) -> aai.LemurModel:
+        if model_name == "claude3_5_sonnet":
+            return aai.LemurModel.claude3_5_sonnet
+        if model_name == "claude3_opus":
+            return aai.LemurModel.claude3_opus
+        if model_name == "claude3_haiku":
+            return aai.LemurModel.claude3_haiku
+        if model_name == "claude3_sonnet":
+            return aai.LemurModel.claude3_sonnet
+        msg = f"Model name not supported: {model_name}"
+        raise ValueError(msg)

lfx/components/assemblyai/assemblyai_list_transcripts.py

@@ -0,0 +1,95 @@
+import assemblyai as aai
+
+from lfx.custom.custom_component.component import Component
+from lfx.io import BoolInput, DropdownInput, IntInput, MessageTextInput, Output, SecretStrInput
+from lfx.log.logger import logger
+from lfx.schema.data import Data
+
+
+class AssemblyAIListTranscripts(Component):
+    display_name = "AssemblyAI List Transcripts"
+    description = "Retrieve a list of transcripts from AssemblyAI with filtering options"
+    documentation = "https://www.assemblyai.com/docs"
+    icon = "AssemblyAI"
+
+    inputs = [
+        SecretStrInput(
+            name="api_key",
+            display_name="Assembly API Key",
+            info="Your AssemblyAI API key. You can get one from https://www.assemblyai.com/",
+            required=True,
+        ),
+        IntInput(
+            name="limit",
+            display_name="Limit",
+            info="Maximum number of transcripts to retrieve (default: 20, use 0 for all)",
+            value=20,
+        ),
+        DropdownInput(
+            name="status_filter",
+            display_name="Status Filter",
+            options=["all", "queued", "processing", "completed", "error"],
+            value="all",
+            info="Filter by transcript status",
+            advanced=True,
+        ),
+        MessageTextInput(
+            name="created_on",
+            display_name="Created On",
+            info="Only get transcripts created on this date (YYYY-MM-DD)",
+            advanced=True,
+        ),
+        BoolInput(
+            name="throttled_only",
+            display_name="Throttled Only",
+            info="Only get throttled transcripts, overrides the status filter",
+            advanced=True,
+        ),
+    ]
+
+    outputs = [
+        Output(display_name="Transcript List", name="transcript_list", method="list_transcripts"),
+    ]
+
+    def list_transcripts(self) -> list[Data]:
+        aai.settings.api_key = self.api_key
+
+        params = aai.ListTranscriptParameters()
+        if self.limit:
+            params.limit = self.limit
+        if self.status_filter != "all":
+            params.status = self.status_filter
+        if self.created_on and self.created_on.text:
+            params.created_on = self.created_on.text
+        if self.throttled_only:
+            params.throttled_only = True
+
+        try:
+            transcriber = aai.Transcriber()
+
+            def convert_page_to_data_list(page):
+                return [Data(**t.dict()) for t in page.transcripts]
+
+            if self.limit == 0:
+                # paginate over all pages
+                params.limit = 100
+                page = transcriber.list_transcripts(params)
+                transcripts = convert_page_to_data_list(page)
+
+                while page.page_details.before_id_of_prev_url is not None:
+                    params.before_id = page.page_details.before_id_of_prev_url
+                    page = transcriber.list_transcripts(params)
+                    transcripts.extend(convert_page_to_data_list(page))
+            else:
+                # just one page
+                page = transcriber.list_transcripts(params)
+                transcripts = convert_page_to_data_list(page)
+
+        except Exception as e:  # noqa: BLE001
+            logger.debug("Error listing transcripts", exc_info=True)
+            error_data = Data(data={"error": f"An error occurred: {e}"})
+            self.status = [error_data]
+            return [error_data]
+
+        self.status = transcripts
+        return transcripts

lfx/components/assemblyai/assemblyai_poll_transcript.py

@@ -0,0 +1,72 @@
+import assemblyai as aai
+
+from lfx.custom.custom_component.component import Component
+from lfx.field_typing.range_spec import RangeSpec
+from lfx.io import DataInput, FloatInput, Output, SecretStrInput
+from lfx.log.logger import logger
+from lfx.schema.data import Data
+
+
+class AssemblyAITranscriptionJobPoller(Component):
+    display_name = "AssemblyAI Poll Transcript"
+    description = "Poll for the status of a transcription job using AssemblyAI"
+    documentation = "https://www.assemblyai.com/docs"
+    icon = "AssemblyAI"
+
+    inputs = [
+        SecretStrInput(
+            name="api_key",
+            display_name="Assembly API Key",
+            info="Your AssemblyAI API key. You can get one from https://www.assemblyai.com/",
+            required=True,
+        ),
+        DataInput(
+            name="transcript_id",
+            display_name="Transcript ID",
+            info="The ID of the transcription job to poll",
+            required=True,
+        ),
+        FloatInput(
+            name="polling_interval",
+            display_name="Polling Interval",
+            value=3.0,
+            info="The polling interval in seconds",
+            advanced=True,
+            range_spec=RangeSpec(min=3, max=30),
+        ),
+    ]
+
+    outputs = [
+        Output(display_name="Transcription Result", name="transcription_result", method="poll_transcription_job"),
+    ]
+
+    def poll_transcription_job(self) -> Data:
+        """Polls the transcription status until completion and returns the Data."""
+        aai.settings.api_key = self.api_key
+        aai.settings.polling_interval = self.polling_interval
+
+        # check if it's an error message from the previous step
+        if self.transcript_id.data.get("error"):
+            self.status = self.transcript_id.data["error"]
+            return self.transcript_id
+
+        try:
+            transcript = aai.Transcript.get_by_id(self.transcript_id.data["transcript_id"])
+        except Exception as e:  # noqa: BLE001
+            error = f"Getting transcription failed: {e}"
+            logger.debug(error, exc_info=True)
+            self.status = error
+            return Data(data={"error": error})
+
+        if transcript.status == aai.TranscriptStatus.completed:
+            json_response = transcript.json_response
+            text = json_response.pop("text", None)
+            utterances = json_response.pop("utterances", None)
+            transcript_id = json_response.pop("id", None)
+            sorted_data = {"text": text, "utterances": utterances, "id": transcript_id}
+            sorted_data.update(json_response)
+            data = Data(data=sorted_data)
+            self.status = data
+            return data
+        self.status = transcript.error
+        return Data(data={"error": transcript.error})

lfx/components/assemblyai/assemblyai_start_transcript.py

@@ -0,0 +1,188 @@
+from pathlib import Path
+
+import assemblyai as aai
+
+from lfx.custom.custom_component.component import Component
+from lfx.io import BoolInput, DropdownInput, FileInput, MessageTextInput, Output, SecretStrInput
+from lfx.log.logger import logger
+from lfx.schema.data import Data
+
+
+class AssemblyAITranscriptionJobCreator(Component):
+    display_name = "AssemblyAI Start Transcript"
+    description = "Create a transcription job for an audio file using AssemblyAI with advanced options"
+    documentation = "https://www.assemblyai.com/docs"
+    icon = "AssemblyAI"
+
+    inputs = [
+        SecretStrInput(
+            name="api_key",
+            display_name="Assembly API Key",
+            info="Your AssemblyAI API key. You can get one from https://www.assemblyai.com/",
+            required=True,
+        ),
+        FileInput(
+            name="audio_file",
+            display_name="Audio File",
+            file_types=[
+                "3ga",
+                "8svx",
+                "aac",
+                "ac3",
+                "aif",
+                "aiff",
+                "alac",
+                "amr",
+                "ape",
+                "au",
+                "dss",
+                "flac",
+                "flv",
+                "m4a",
+                "m4b",
+                "m4p",
+                "m4r",
+                "mp3",
+                "mpga",
+                "ogg",
+                "oga",
+                "mogg",
+                "opus",
+                "qcp",
+                "tta",
+                "voc",
+                "wav",
+                "wma",
+                "wv",
+                "webm",
+                "mts",
+                "m2ts",
+                "ts",
+                "mov",
+                "mp2",
+                "mp4",
+                "m4p",
+                "m4v",
+                "mxf",
+            ],
+            info="The audio file to transcribe",
+            required=True,
+        ),
+        MessageTextInput(
+            name="audio_file_url",
+            display_name="Audio File URL",
+            info="The URL of the audio file to transcribe (Can be used instead of a File)",
+            advanced=True,
+        ),
+        DropdownInput(
+            name="speech_model",
+            display_name="Speech Model",
+            options=[
+                "best",
+                "nano",
+            ],
+            value="best",
+            info="The speech model to use for the transcription",
+            advanced=True,
+        ),
+        BoolInput(
+            name="language_detection",
+            display_name="Automatic Language Detection",
+            info="Enable automatic language detection",
+            advanced=True,
+        ),
+        MessageTextInput(
+            name="language_code",
+            display_name="Language",
+            info=(
+                """
+            The language of the audio file. Can be set manually if automatic language detection is disabled.
+            See https://www.assemblyai.com/docs/getting-started/supported-languages """
+                "for a list of supported language codes."
+            ),
+            advanced=True,
+        ),
+        BoolInput(
+            name="speaker_labels",
+            display_name="Enable Speaker Labels",
+            info="Enable speaker diarization",
+        ),
+        MessageTextInput(
+            name="speakers_expected",
+            display_name="Expected Number of Speakers",
+            info="Set the expected number of speakers (optional, enter a number)",
+            advanced=True,
+        ),
+        BoolInput(
+            name="punctuate",
+            display_name="Punctuate",
+            info="Enable automatic punctuation",
+            advanced=True,
+            value=True,
+        ),
+        BoolInput(
+            name="format_text",
+            display_name="Format Text",
+            info="Enable text formatting",
+            advanced=True,
+            value=True,
+        ),
+    ]
+
+    outputs = [
+        Output(display_name="Transcript ID", name="transcript_id", method="create_transcription_job"),
+    ]
+
+    def create_transcription_job(self) -> Data:
+        aai.settings.api_key = self.api_key
+
+        # Convert speakers_expected to int if it's not empty
+        speakers_expected = None
+        if self.speakers_expected and self.speakers_expected.strip():
+            try:
+                speakers_expected = int(self.speakers_expected)
+            except ValueError:
+                self.status = "Error: Expected Number of Speakers must be a valid integer"
+                return Data(data={"error": "Error: Expected Number of Speakers must be a valid integer"})
+
+        language_code = self.language_code or None
+
+        config = aai.TranscriptionConfig(
+            speech_model=self.speech_model,
+            language_detection=self.language_detection,
+            language_code=language_code,
+            speaker_labels=self.speaker_labels,
+            speakers_expected=speakers_expected,
+            punctuate=self.punctuate,
+            format_text=self.format_text,
+        )
+
+        audio = None
+        if self.audio_file:
+            if self.audio_file_url:
+                logger.warning("Both an audio file an audio URL were specified. The audio URL was ignored.")
+
+            # Check if the file exists
+            if not Path(self.audio_file).exists():
+                self.status = "Error: Audio file not found"
+                return Data(data={"error": "Error: Audio file not found"})
+            audio = self.audio_file
+        elif self.audio_file_url:
+            audio = self.audio_file_url
+        else:
+            self.status = "Error: Either an audio file or an audio URL must be specified"
+            return Data(data={"error": "Error: Either an audio file or an audio URL must be specified"})
+
+        try:
+            transcript = aai.Transcriber().submit(audio, config=config)
+        except Exception as e:  # noqa: BLE001
+            logger.debug("Error submitting transcription job", exc_info=True)
+            self.status = f"An error occurred: {e}"
+            return Data(data={"error": f"An error occurred: {e}"})
+
+        if transcript.error:
+            self.status = transcript.error
+            return Data(data={"error": transcript.error})
+        result = Data(data={"transcript_id": transcript.id})
+        self.status = result
+        return result

lfx/components/azure/__init__.py

@@ -0,0 +1,37 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any
+
+from lfx.components._importing import import_mod
+
+if TYPE_CHECKING:
+    from .azure_openai import AzureChatOpenAIComponent
+    from .azure_openai_embeddings import AzureOpenAIEmbeddingsComponent
+
+_dynamic_imports = {
+    "AzureChatOpenAIComponent": "azure_openai",
+    "AzureOpenAIEmbeddingsComponent": "azure_openai_embeddings",
+}
+
+__all__ = [
+    "AzureChatOpenAIComponent",
+    "AzureOpenAIEmbeddingsComponent",
+]
+
+
+def __getattr__(attr_name: str) -> Any:
+    """Lazily import azure components on attribute access."""
+    if attr_name not in _dynamic_imports:
+        msg = f"module '{__name__}' has no attribute '{attr_name}'"
+        raise AttributeError(msg)
+    try:
+        result = import_mod(attr_name, _dynamic_imports[attr_name], __spec__.parent)
+    except (ModuleNotFoundError, ImportError, AttributeError) as e:
+        msg = f"Could not import '{attr_name}' from '{__name__}': {e}"
+        raise AttributeError(msg) from e
+    globals()[attr_name] = result
+    return result
+
+
+def __dir__() -> list[str]:
+    return list(__all__)

lfx/components/azure/azure_openai.py

@@ -0,0 +1,95 @@
+from langchain_openai import AzureChatOpenAI
+
+from lfx.base.models.model import LCModelComponent
+from lfx.field_typing import LanguageModel
+from lfx.field_typing.range_spec import RangeSpec
+from lfx.inputs.inputs import MessageTextInput
+from lfx.io import DropdownInput, IntInput, SecretStrInput, SliderInput
+
+
+class AzureChatOpenAIComponent(LCModelComponent):
+    display_name: str = "Azure OpenAI"
+    description: str = "Generate text using Azure OpenAI LLMs."
+    documentation: str = "https://python.langchain.com/docs/integrations/llms/azure_openai"
+    beta = False
+    icon = "Azure"
+    name = "AzureOpenAIModel"
+
+    AZURE_OPENAI_API_VERSIONS = [
+        "2024-06-01",
+        "2024-07-01-preview",
+        "2024-08-01-preview",
+        "2024-09-01-preview",
+        "2024-10-01-preview",
+        "2023-05-15",
+        "2023-12-01-preview",
+        "2024-02-15-preview",
+        "2024-03-01-preview",
+        "2024-12-01-preview",
+        "2025-01-01-preview",
+        "2025-02-01-preview",
+    ]
+
+    inputs = [
+        *LCModelComponent.get_base_inputs(),
+        MessageTextInput(
+            name="azure_endpoint",
+            display_name="Azure Endpoint",
+            info="Your Azure endpoint, including the resource. Example: `https://example-resource.azure.openai.com/`",
+            required=True,
+        ),
+        MessageTextInput(name="azure_deployment", display_name="Deployment Name", required=True),
+        SecretStrInput(name="api_key", display_name="Azure Chat OpenAI API Key", required=True),
+        DropdownInput(
+            name="api_version",
+            display_name="API Version",
+            options=sorted(AZURE_OPENAI_API_VERSIONS, reverse=True),
+            value=next(
+                (
+                    version
+                    for version in sorted(AZURE_OPENAI_API_VERSIONS, reverse=True)
+                    if not version.endswith("-preview")
+                ),
+                AZURE_OPENAI_API_VERSIONS[0],
+            ),
+        ),
+        SliderInput(
+            name="temperature",
+            display_name="Temperature",
+            value=0.7,
+            range_spec=RangeSpec(min=0, max=2, step=0.01),
+            info="Controls randomness. Lower values are more deterministic, higher values are more creative.",
+            advanced=True,
+        ),
+        IntInput(
+            name="max_tokens",
+            display_name="Max Tokens",
+            advanced=True,
+            info="The maximum number of tokens to generate. Set to 0 for unlimited tokens.",
+        ),
+    ]
+
+    def build_model(self) -> LanguageModel:  # type: ignore[type-var]
+        azure_endpoint = self.azure_endpoint
+        azure_deployment = self.azure_deployment
+        api_version = self.api_version
+        api_key = self.api_key
+        temperature = self.temperature
+        max_tokens = self.max_tokens
+        stream = self.stream
+
+        try:
+            output = AzureChatOpenAI(
+                azure_endpoint=azure_endpoint,
+                azure_deployment=azure_deployment,
+                api_version=api_version,
+                api_key=api_key,
+                temperature=temperature,
+                max_tokens=max_tokens or None,
+                streaming=stream,
+            )
+        except Exception as e:
+            msg = f"Could not connect to AzureOpenAI API: {e}"
+            raise ValueError(msg) from e
+
+        return output