@weirdfingers/baseboards 0.9.5 → 0.9.7

package/templates/api/src/boards/generators/implementations/fal/audio/chatterbox_tts_turbo.py

@@ -1,195 +0,0 @@
-"""
-fal.ai Chatterbox Text-to-Speech Turbo generator.
-
-Generate expressive speech from text with paralinguistic controls like laughs,
-sighs, coughs, and more. Supports voice cloning with custom audio samples.
-
-Based on Fal AI's fal-ai/chatterbox/text-to-speech/turbo model.
-See: https://fal.ai/models/fal-ai/chatterbox/text-to-speech/turbo
-"""
-
-import os
-from typing import Literal
-
-from pydantic import BaseModel, Field
-
-from ....artifacts import AudioArtifact
-from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
-
-# Voice presets available in Chatterbox
-ChatterboxVoice = Literal[
-    "aaron",
-    "abigail",
-    "anaya",
-    "andy",
-    "archer",
-    "brian",
-    "chloe",
-    "dylan",
-    "emmanuel",
-    "ethan",
-    "evelyn",
-    "gavin",
-    "gordon",
-    "ivan",
-    "laura",
-    "lucy",
-    "madison",
-    "marisol",
-    "meera",
-    "walter",
-]
-
-
-class ChatterboxTtsTurboInput(BaseModel):
-    """Input schema for Chatterbox TTS Turbo.
-
-    Artifact fields are automatically detected via type introspection
-    and resolved from generation IDs to artifact objects.
-    """
-
-    text: str = Field(
-        description=(
-            "The text to be converted to speech. Supports paralinguistic tags: "
-            "[clear throat], [sigh], [shush], [cough], [groan], [sniff], [gasp], "
-            "[chuckle], [laugh]"
-        ),
-        min_length=1,
-    )
-    voice: ChatterboxVoice = Field(
-        default="lucy",
-        description="Preset voice to use for synthesis",
-    )
-    audio_url: AudioArtifact | None = Field(
-        default=None,
-        description=(
-            "Optional audio file (5-10 seconds) for voice cloning. "
-            "If provided, this overrides the preset voice selection."
-        ),
-    )
-    temperature: float = Field(
-        default=0.8,
-        ge=0.05,
-        le=2.0,
-        description="Temperature for generation. Higher values create more varied speech patterns.",
-    )
-    seed: int | None = Field(
-        default=None,
-        description="Random seed for reproducible results. Set for consistent generations.",
-    )
-
-
-class FalChatterboxTtsTurboGenerator(BaseGenerator):
-    """Chatterbox TTS Turbo text-to-speech generator using fal.ai."""
-
-    name = "fal-chatterbox-tts-turbo"
-    artifact_type = "audio"
-    description = (
-        "Fal: Chatterbox TTS Turbo - "
-        "Expressive text-to-speech with paralinguistic controls and voice cloning"
-    )
-
-    def get_input_schema(self) -> type[ChatterboxTtsTurboInput]:
-        return ChatterboxTtsTurboInput
-
-    async def generate(
-        self, inputs: ChatterboxTtsTurboInput, context: GeneratorExecutionContext
-    ) -> GeneratorResult:
-        """Generate audio using fal.ai Chatterbox TTS Turbo model."""
-        # Check for API key (fal-client uses FAL_KEY environment variable)
-        if not os.getenv("FAL_KEY"):
-            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
-
-        # Import fal_client
-        try:
-            import fal_client
-        except ImportError as e:
-            raise ImportError(
-                "fal.ai SDK is required for FalChatterboxTtsTurboGenerator. "
-                "Install with: pip install weirdfingers-boards[generators-fal]"
-            ) from e
-
-        # Prepare arguments for fal.ai API
-        arguments: dict[str, str | float | int] = {
-            "text": inputs.text,
-            "voice": inputs.voice,
-            "temperature": inputs.temperature,
-        }
-
-        # Add seed if provided
-        if inputs.seed is not None:
-            arguments["seed"] = inputs.seed
-
-        # Handle voice cloning audio upload
-        if inputs.audio_url is not None:
-            from ..utils import upload_artifacts_to_fal
-
-            audio_urls = await upload_artifacts_to_fal([inputs.audio_url], context)
-            arguments["audio_url"] = audio_urls[0]
-
-        # Submit async job and get handler
-        handler = await fal_client.submit_async(
-            "fal-ai/chatterbox/text-to-speech/turbo",
-            arguments=arguments,
-        )
-
-        # Store the external job ID for tracking
-        await context.set_external_job_id(handler.request_id)
-
-        # Stream progress updates (sample every 3rd event to avoid spam)
-        from .....progress.models import ProgressUpdate
-
-        event_count = 0
-        async for event in handler.iter_events(with_logs=True):
-            event_count += 1
-
-            # Process every 3rd event to provide feedback without overwhelming
-            if event_count % 3 == 0:
-                # Extract logs if available
-                logs = getattr(event, "logs", None)
-                if logs:
-                    # Join log entries into a single message
-                    if isinstance(logs, list):
-                        message = " | ".join(str(log) for log in logs if log)
-                    else:
-                        message = str(logs)
-
-                    if message:
-                        await context.publish_progress(
-                            ProgressUpdate(
-                                job_id=handler.request_id,
-                                status="processing",
-                                progress=50.0,  # Approximate mid-point progress
-                                phase="processing",
-                                message=message,
-                            )
-                        )
-
-        # Get final result
-        result = await handler.get()
-
-        # Extract audio URL from result
-        # fal.ai returns: {"audio": {"url": "..."}}
-        audio_data = result.get("audio")
-        if audio_data is None:
-            raise ValueError("No audio data returned from fal.ai API")
-
-        audio_url = audio_data.get("url")
-        if not audio_url:
-            raise ValueError("Audio URL missing in fal.ai response")
-
-        # Store audio result
-        artifact = await context.store_audio_result(
-            storage_url=audio_url,
-            format="wav",  # Chatterbox TTS returns WAV format
-            output_index=0,
-        )
-
-        return GeneratorResult(outputs=[artifact])
-
-    async def estimate_cost(self, inputs: ChatterboxTtsTurboInput) -> float:
-        """Estimate cost for Chatterbox TTS Turbo generation.
-
-        Chatterbox TTS Turbo pricing is approximately $0.03 per generation.
-        """
-        return 0.03

package/templates/api/src/boards/generators/implementations/fal/audio/elevenlabs_sound_effects_v2.py

@@ -1,194 +0,0 @@
-"""
-fal.ai ElevenLabs Sound Effects V2 text-to-audio generator.
-
-Generate sound effects from text descriptions using ElevenLabs advanced
-sound effects model. Supports customizable duration, prompt influence,
-and multiple output formats.
-
-Based on Fal AI's fal-ai/elevenlabs/sound-effects/v2 model.
-See: https://fal.ai/models/fal-ai/elevenlabs/sound-effects/v2
-"""
-
-import os
-from typing import Literal
-
-from pydantic import BaseModel, Field
-
-from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
-
-
-class ElevenlabsSoundEffectsV2Input(BaseModel):
-    """Input schema for ElevenLabs Sound Effects V2 generation.
-
-    Generates custom sound effects from natural language descriptions with
-    configurable duration, prompt influence, and output format.
-    """
-
-    text: str = Field(
-        description="Text describing the sound effect to generate",
-        examples=[
-            "Spacious braam suitable for high-impact movie trailer moments",
-            "A gentle wind chime tinkling in a soft breeze",
-        ],
-    )
-    duration_seconds: float | None = Field(
-        default=None,
-        ge=0.5,
-        le=22.0,
-        description=(
-            "Duration in seconds (0.5-22). "
-            "If None, optimal duration will be determined from prompt."
-        ),
-    )
-    prompt_influence: float = Field(
-        default=0.3,
-        ge=0.0,
-        le=1.0,
-        description="How closely to follow the prompt (0-1). Higher values mean less variation.",
-    )
-    output_format: Literal[
-        "mp3_22050_32",
-        "mp3_44100_32",
-        "mp3_44100_64",
-        "mp3_44100_96",
-        "mp3_44100_128",
-        "mp3_44100_192",
-        "pcm_8000",
-        "pcm_16000",
-        "pcm_22050",
-        "pcm_24000",
-        "pcm_44100",
-        "pcm_48000",
-        "ulaw_8000",
-        "alaw_8000",
-        "opus_48000_32",
-        "opus_48000_64",
-        "opus_48000_96",
-        "opus_48000_128",
-        "opus_48000_192",
-    ] = Field(
-        default="mp3_44100_128",
-        description=(
-            "Output format of the generated audio. Formatted as codec_sample_rate_bitrate."
-        ),
-    )
-    loop: bool = Field(
-        default=False,
-        description="Whether to create a sound effect that loops smoothly.",
-    )
-
-
-class FalElevenlabsSoundEffectsV2Generator(BaseGenerator):
-    """ElevenLabs Sound Effects V2 text-to-audio generator using fal.ai."""
-
-    name = "fal-elevenlabs-sound-effects-v2"
-    artifact_type = "audio"
-    description = (
-        "Fal: ElevenLabs Sound Effects V2 - "
-        "Generate custom sound effects from text descriptions with advanced AI"
-    )
-
-    def get_input_schema(self) -> type[ElevenlabsSoundEffectsV2Input]:
-        return ElevenlabsSoundEffectsV2Input
-
-    async def generate(
-        self, inputs: ElevenlabsSoundEffectsV2Input, context: GeneratorExecutionContext
-    ) -> GeneratorResult:
-        """Generate sound effect using fal.ai ElevenLabs Sound Effects V2 model."""
-        # Check for API key (fal-client uses FAL_KEY environment variable)
-        if not os.getenv("FAL_KEY"):
-            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
-
-        # Import fal_client
-        try:
-            import fal_client
-        except ImportError as e:
-            raise ImportError(
-                "fal.ai SDK is required for FalElevenlabsSoundEffectsV2Generator. "
-                "Install with: pip install weirdfingers-boards[generators-fal]"
-            ) from e
-
-        # Prepare arguments for fal.ai API
-        arguments = {
-            "text": inputs.text,
-            "prompt_influence": inputs.prompt_influence,
-            "output_format": inputs.output_format,
-            "loop": inputs.loop,
-        }
-
-        # Only include duration_seconds if specified
-        if inputs.duration_seconds is not None:
-            arguments["duration_seconds"] = inputs.duration_seconds
-
-        # Submit async job and get handler
-        handler = await fal_client.submit_async(
-            "fal-ai/elevenlabs/sound-effects/v2",
-            arguments=arguments,
-        )
-
-        # Store the external job ID for tracking
-        await context.set_external_job_id(handler.request_id)
-
-        # Stream progress updates (sample every 3rd event to avoid spam)
-        from .....progress.models import ProgressUpdate
-
-        event_count = 0
-        async for event in handler.iter_events(with_logs=True):
-            event_count += 1
-
-            # Process every 3rd event to provide feedback without overwhelming
-            if event_count % 3 == 0:
-                # Extract logs if available
-                logs = getattr(event, "logs", None)
-                if logs:
-                    # Join log entries into a single message
-                    if isinstance(logs, list):
-                        message = " | ".join(str(log) for log in logs if log)
-                    else:
-                        message = str(logs)
-
-                    if message:
-                        await context.publish_progress(
-                            ProgressUpdate(
-                                job_id=handler.request_id,
-                                status="processing",
-                                progress=50.0,  # Approximate mid-point progress
-                                phase="processing",
-                                message=message,
-                            )
-                        )
-
-        # Get final result
-        result = await handler.get()
-
-        # Extract audio URL from result
-        # fal.ai returns: {"audio": {"url": "...", "content_type": "...", ...}}
-        audio_data = result.get("audio")
-        if audio_data is None:
-            raise ValueError("No audio data returned from fal.ai API")
-
-        audio_url = audio_data.get("url")
-        if not audio_url:
-            raise ValueError("Audio URL missing in fal.ai response")
-
-        # Extract format from output_format (e.g., "mp3_44100_128" -> "mp3")
-        format_parts = inputs.output_format.split("_")
-        audio_format = format_parts[0] if format_parts else "mp3"
-
-        # Store audio result
-        artifact = await context.store_audio_result(
-            storage_url=audio_url,
-            format=audio_format,
-            output_index=0,
-        )
-
-        return GeneratorResult(outputs=[artifact])
-
-    async def estimate_cost(self, inputs: ElevenlabsSoundEffectsV2Input) -> float:
-        """Estimate cost for ElevenLabs Sound Effects V2 generation.
-
-        Based on typical ElevenLabs sound effects pricing.
-        Cost is approximately $0.055 per sound effect generation.
-        """
-        # Fixed cost per generation regardless of duration or format
-        return 0.055

package/templates/api/src/boards/generators/implementations/fal/audio/elevenlabs_tts_eleven_v3.py

@@ -1,209 +0,0 @@
-"""
-fal.ai ElevenLabs Text-to-Speech Eleven-V3 generator.
-
-Generate high-quality speech from text using ElevenLabs' Eleven-V3 model,
-offering natural-sounding voices with customizable parameters for stability,
-similarity, style, and speed.
-
-Based on Fal AI's fal-ai/elevenlabs/tts/eleven-v3 model.
-See: https://fal.ai/models/fal-ai/elevenlabs/tts/eleven-v3
-"""
-
-import os
-
-from pydantic import BaseModel, Field
-
-from ....base import BaseGenerator, GeneratorExecutionContext, GeneratorResult
-
-
-class ElevenlabsTtsElevenV3Input(BaseModel):
-    """Input schema for ElevenLabs TTS Eleven-V3 generation.
-
-    The text is converted to speech using advanced AI voice synthesis with
-    customizable voice characteristics and optional word-level timestamps.
-    """
-
-    text: str = Field(
-        description="The text to convert to speech",
-        min_length=1,
-    )
-    voice: str = Field(
-        default="Rachel",
-        description=(
-            "Voice selection. Available voices: "
-            "Aria, Roger, Sarah, Laura, Charlie, George, Callum, River, Liam, "
-            "Charlotte, Alice, Matilda, Will, Jessica, Eric, Chris, Brian, "
-            "Daniel, Lily, Bill, Rachel"
-        ),
-    )
-    stability: float = Field(
-        default=0.5,
-        ge=0.0,
-        le=1.0,
-        description="Voice stability. Higher values result in more consistent output",
-    )
-    similarity_boost: float = Field(
-        default=0.75,
-        ge=0.0,
-        le=1.0,
-        description="Similarity boost for voice matching",
-    )
-    style: float = Field(
-        default=0.0,
-        ge=0.0,
-        le=1.0,
-        description="Style exaggeration. Higher values add more expressiveness",
-    )
-    speed: float = Field(
-        default=1.0,
-        ge=0.7,
-        le=1.2,
-        description="Speech rate adjustment. 1.0 is normal speed",
-    )
-    timestamps: bool = Field(
-        default=False,
-        description="Whether to return timestamps for each word",
-    )
-    previous_text: str | None = Field(
-        default=None,
-        description="Context from prior content for improved continuity",
-    )
-    next_text: str | None = Field(
-        default=None,
-        description="Context for upcoming content for improved continuity",
-    )
-    language_code: str | None = Field(
-        default=None,
-        description="ISO 639-1 language code (limited model support)",
-    )
-
-
-class FalElevenlabsTtsElevenV3Generator(BaseGenerator):
-    """ElevenLabs Text-to-Speech Eleven-V3 generator using fal.ai."""
-
-    name = "fal-elevenlabs-tts-eleven-v3"
-    artifact_type = "audio"
-    description = (
-        "Fal: ElevenLabs TTS Eleven-V3 - "
-        "High-quality text-to-speech with natural-sounding voices and customizable parameters"
-    )
-
-    def get_input_schema(self) -> type[ElevenlabsTtsElevenV3Input]:
-        return ElevenlabsTtsElevenV3Input
-
-    async def generate(
-        self, inputs: ElevenlabsTtsElevenV3Input, context: GeneratorExecutionContext
-    ) -> GeneratorResult:
-        """Generate audio using fal.ai ElevenLabs TTS Eleven-V3 model."""
-        # Check for API key (fal-client uses FAL_KEY environment variable)
-        if not os.getenv("FAL_KEY"):
-            raise ValueError("API configuration invalid. Missing FAL_KEY environment variable")
-
-        # Import fal_client
-        try:
-            import fal_client
-        except ImportError as e:
-            raise ImportError(
-                "fal.ai SDK is required for FalElevenlabsTtsElevenV3Generator. "
-                "Install with: pip install weirdfingers-boards[generators-fal]"
-            ) from e
-
-        # Prepare arguments for fal.ai API
-        arguments = {
-            "text": inputs.text,
-            "voice": inputs.voice,
-            "stability": inputs.stability,
-            "similarity_boost": inputs.similarity_boost,
-            "style": inputs.style,
-            "speed": inputs.speed,
-            "timestamps": inputs.timestamps,
-        }
-
-        # Add optional context parameters if provided
-        if inputs.previous_text is not None:
-            arguments["previous_text"] = inputs.previous_text
-        if inputs.next_text is not None:
-            arguments["next_text"] = inputs.next_text
-        if inputs.language_code is not None:
-            arguments["language_code"] = inputs.language_code
-
-        # Submit async job and get handler
-        handler = await fal_client.submit_async(
-            "fal-ai/elevenlabs/tts/eleven-v3",
-            arguments=arguments,
-        )
-
-        # Store the external job ID for tracking
-        await context.set_external_job_id(handler.request_id)
-
-        # Stream progress updates (sample every 3rd event to avoid spam)
-        from .....progress.models import ProgressUpdate
-
-        event_count = 0
-        async for event in handler.iter_events(with_logs=True):
-            event_count += 1
-
-            # Process every 3rd event to provide feedback without overwhelming
-            if event_count % 3 == 0:
-                # Extract logs if available
-                logs = getattr(event, "logs", None)
-                if logs:
-                    # Join log entries into a single message
-                    if isinstance(logs, list):
-                        message = " | ".join(str(log) for log in logs if log)
-                    else:
-                        message = str(logs)
-
-                    if message:
-                        await context.publish_progress(
-                            ProgressUpdate(
-                                job_id=handler.request_id,
-                                status="processing",
-                                progress=50.0,  # Approximate mid-point progress
-                                phase="processing",
-                                message=message,
-                            )
-                        )
-
-        # Get final result
-        result = await handler.get()
-
-        # Extract audio URL from result
-        # fal.ai returns: {"audio": {"url": "...", "content_type": "...", ...}}
-        audio_data = result.get("audio")
-        if audio_data is None:
-            raise ValueError("No audio data returned from fal.ai API")
-
-        audio_url = audio_data.get("url")
-        if not audio_url:
-            raise ValueError("Audio URL missing in fal.ai response")
-
-        # Determine format from content_type or default to mp3
-        content_type = audio_data.get("content_type", "audio/mpeg")
-        format_map = {
-            "audio/mpeg": "mp3",
-            "audio/mp3": "mp3",
-            "audio/wav": "wav",
-            "audio/ogg": "ogg",
-        }
-        audio_format = format_map.get(content_type, "mp3")
-
-        # Store audio result
-        artifact = await context.store_audio_result(
-            storage_url=audio_url,
-            format=audio_format,
-            output_index=0,
-        )
-
-        return GeneratorResult(outputs=[artifact])
-
-    async def estimate_cost(self, inputs: ElevenlabsTtsElevenV3Input) -> float:
-        """Estimate cost for ElevenLabs TTS Eleven-V3 generation.
-
-        ElevenLabs TTS Eleven-V3 costs $0.10 per 1000 characters.
-        """
-        # Calculate character count
-        char_count = len(inputs.text)
-
-        # Cost is $0.10 per 1000 characters
-        return (char_count / 1000.0) * 0.10