videosdk-plugins-resemble 0.0.17__py3-none-any.whl

videosdk/plugins/resemble/__init__.py

@@ -0,0 +1,3 @@
+from .tts import ResembleTTS
+
+__all__ = ["ResembleTTS"]

videosdk/plugins/resemble/tts.py

@@ -0,0 +1,114 @@
+from __future__ import annotations
+
+from typing import Any, AsyncIterator, Optional
+import os
+import asyncio
+import httpx
+from dataclasses import dataclass
+
+from videosdk.agents import TTS
+
+RESEMBLE_HTTP_STREAMING_URL = "https://f.cluster.resemble.ai/stream"
+DEFAULT_VOICE_UUID = "55592656"
+DEFAULT_SAMPLE_RATE = 22050
+DEFAULT_PRECISION = "PCM_16"
+
+class ResembleTTS(TTS):
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        voice_uuid: str = DEFAULT_VOICE_UUID,
+        sample_rate: int = DEFAULT_SAMPLE_RATE,
+        precision: str = DEFAULT_PRECISION,
+    ) -> None:
+        super().__init__(sample_rate=sample_rate, num_channels=1)
+
+        self.api_key = api_key or os.getenv("RESEMBLE_API_KEY")
+        if not self.api_key:
+            raise ValueError("Resemble API key is required. Provide either `api_key` or set `RESEMBLE_API_KEY` environment variable.")
+        
+        self.voice_uuid = voice_uuid
+        self.precision = precision
+
+        self.audio_track = None
+        self.loop = None
+        self._http_client = httpx.AsyncClient(
+            timeout=httpx.Timeout(connect=15.0, read=30.0, write=5.0, pool=5.0),
+            follow_redirects=True,
+        )
+
+    async def synthesize(
+        self,
+        text: AsyncIterator[str] | str,
+        **kwargs: Any,
+    ) -> None:
+        try:
+            if isinstance(text, AsyncIterator):
+                full_text = ""
+                async for chunk in text:
+                    full_text += chunk
+            else:
+                full_text = text
+
+            if not self.audio_track or not self.loop:
+                self.emit("error", "Audio track or event loop not set")
+                return
+
+            await self._http_stream_synthesis(full_text)
+
+        except Exception as e:
+            self.emit("error", f"Resemble TTS synthesis failed: {str(e)}")
+
+    async def _http_stream_synthesis(self, text: str) -> None:
+        headers = {
+            "Authorization": f"Token {self.api_key}",
+            "Content-Type": "application/json",
+        }
+        
+        payload = {
+            "voice_uuid": self.voice_uuid,
+            "data": text,
+            "precision": self.precision,
+            "sample_rate": self.sample_rate,
+        }
+
+        try:
+            async with self._http_client.stream(
+                "POST", 
+                RESEMBLE_HTTP_STREAMING_URL,
+                headers=headers, 
+                json=payload
+            ) as response:
+                response.raise_for_status()
+
+                header_processed = False
+                buffer = b''
+
+                async for chunk in response.aiter_bytes():
+                    if not header_processed:
+                        buffer += chunk
+                        data_pos = buffer.find(b'data')
+                        if data_pos != -1:
+                            header_size = data_pos + 8
+                            audio_data = buffer[header_size:]
+                            if audio_data:
+                                self.loop.create_task(self.audio_track.add_new_bytes(audio_data))
+                            header_processed = True
+                    else:
+                        if chunk:
+                            self.loop.create_task(self.audio_track.add_new_bytes(chunk))
+                        
+        except httpx.HTTPStatusError as e:
+            self.emit("error", f"HTTP error {e.response.status_code}: {e.response.text}")
+        except Exception as e:
+            self.emit("error", f"HTTP streaming synthesis failed: {str(e)}")
+
+    async def aclose(self) -> None:
+        if self._http_client:
+            await self._http_client.aclose()
+        await super().aclose()
+
+    async def interrupt(self) -> None:
+        if self.audio_track:
+            self.audio_track.interrupt()

videosdk/plugins/resemble/version.py

@@ -0,0 +1 @@
+__version__ = "0.0.17"

videosdk_plugins_resemble-0.0.17.dist-info/METADATA

@@ -0,0 +1,26 @@
+Metadata-Version: 2.4
+Name: videosdk-plugins-resemble
+Version: 0.0.17
+Summary: VideoSDK Agent Framework plugin for Resemble
+Author: videosdk
+License-Expression: Apache-2.0
+Keywords: ai,audio,resemble,video,videosdk
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Communications :: Conferencing
+Classifier: Topic :: Multimedia :: Sound/Audio
+Classifier: Topic :: Multimedia :: Video
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.11
+Requires-Dist: videosdk-agents>=0.0.17
+Description-Content-Type: text/markdown
+
+VideoSDK Resemble Plugin
+
+Agent Framework plugin for tts services from Resemble.
+
+## Installation
+
+```bash
+pip install videosdk-plugins-resemble
+```

videosdk_plugins_resemble-0.0.17.dist-info/RECORD

@@ -0,0 +1,6 @@
+videosdk/plugins/resemble/__init__.py,sha256=V4pLV2GncUWcXNFXea8Gtf7hbRYKBQSLkW2DsyxW2ks,55
+videosdk/plugins/resemble/tts.py,sha256=DplAv3MHqrTONOsNgngoVy6EX0d-Z7YkBhq0ZPPoifY,3819
+videosdk/plugins/resemble/version.py,sha256=Sd6j21A3v31CfSZ7ASugNxkOYAo0W0NUqcHuzfW7jJA,22
+videosdk_plugins_resemble-0.0.17.dist-info/METADATA,sha256=zICeqp8NFeulvwuehUVb3X0SZAMFkZPc3kQXn8GuznQ,765
+videosdk_plugins_resemble-0.0.17.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+videosdk_plugins_resemble-0.0.17.dist-info/RECORD,,

videosdk_plugins_resemble-0.0.17.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.27.0
+Root-Is-Purelib: true
+Tag: py3-none-any