Address PR review: rename to XAIHttpTTSService, add language map, clean up API
- Rename XAITTSService → XAIHttpTTSService and XAITTSSettings → XAIHttpTTSSettings - Add language_to_xai_language() with explicit LANGUAGE_MAP using resolve_language() - Remove deprecated InputParams, params, voice, language init params - Remove XAI_DEFAULT_SAMPLE_RATE and XAI_PCM_CODEC constants; add encoding param - Set sample_rate=None default (picked up from PipelineParams or user) - Use Language.EN enum instead of string "en" for default language - Add changelog/4031.added.md - Add 07e-interruptible-xai.py foundational example - Update 14g-function-calling-grok.py to use XAIHttpTTSService - Register 07e in run-release-evals.py
This commit is contained in:
Nicholas Zhao
committed by
Mark Backman
Mark Backman
parent
02b97035f8
commit
bbd14de9c5
@@ -15,57 +15,77 @@ from typing import AsyncGenerator, Optional
|
||||
|
||||
import aiohttp
|
||||
from loguru import logger
|
||||
from pydantic import BaseModel
|
||||
|
||||
from pipecat.frames.frames import ErrorFrame, Frame, TTSAudioRawFrame
|
||||
from pipecat.services.settings import TTSSettings
|
||||
from pipecat.services.tts_service import TTSService
|
||||
from pipecat.transcriptions.language import Language
|
||||
from pipecat.transcriptions.language import Language, resolve_language
|
||||
from pipecat.utils.tracing.service_decorators import traced_tts
|
||||
|
||||
|
||||
def language_to_xai_language(language: Language) -> Optional[str]:
|
||||
"""Convert a Language enum to xAI language code.
|
||||
|
||||
Args:
|
||||
language: The Language enum value to convert.
|
||||
|
||||
Returns:
|
||||
The corresponding xAI language code, or None if not supported.
|
||||
"""
|
||||
LANGUAGE_MAP = {
|
||||
Language.AR: "ar-EG",
|
||||
Language.AR_EG: "ar-EG",
|
||||
Language.AR_SA: "ar-SA",
|
||||
Language.AR_AE: "ar-AE",
|
||||
Language.BN: "bn",
|
||||
Language.DE: "de",
|
||||
Language.EN: "en",
|
||||
Language.ES: "es-ES",
|
||||
Language.ES_ES: "es-ES",
|
||||
Language.ES_MX: "es-MX",
|
||||
Language.FR: "fr",
|
||||
Language.HI: "hi",
|
||||
Language.ID: "id",
|
||||
Language.IT: "it",
|
||||
Language.JA: "ja",
|
||||
Language.KO: "ko",
|
||||
Language.PT: "pt-PT",
|
||||
Language.PT_BR: "pt-BR",
|
||||
Language.PT_PT: "pt-PT",
|
||||
Language.RU: "ru",
|
||||
Language.TR: "tr",
|
||||
Language.VI: "vi",
|
||||
Language.ZH: "zh",
|
||||
}
|
||||
|
||||
return resolve_language(language, LANGUAGE_MAP, use_base_code=True)
|
||||
|
||||
|
||||
@dataclass
|
||||
class XAITTSSettings(TTSSettings):
|
||||
"""Settings for XAITTSService."""
|
||||
class XAIHttpTTSSettings(TTSSettings):
|
||||
"""Settings for XAIHttpTTSService."""
|
||||
|
||||
pass
|
||||
|
||||
|
||||
class XAITTSService(TTSService):
|
||||
class XAIHttpTTSService(TTSService):
|
||||
"""xAI HTTP text-to-speech service.
|
||||
|
||||
The service requests raw PCM audio so emitted ``TTSAudioRawFrame`` objects
|
||||
match Pipecat's downstream expectations without extra decoding.
|
||||
"""
|
||||
|
||||
Settings = XAITTSSettings
|
||||
Settings = XAIHttpTTSSettings
|
||||
_settings: Settings
|
||||
|
||||
XAI_DEFAULT_SAMPLE_RATE = 24000
|
||||
XAI_PCM_CODEC = "pcm"
|
||||
|
||||
class InputParams(BaseModel):
|
||||
"""Input parameters for xAI TTS configuration.
|
||||
|
||||
.. deprecated:: 0.0.105
|
||||
Use ``settings=XAITTSService.Settings(...)`` instead.
|
||||
|
||||
Parameters:
|
||||
language: Language for speech synthesis.
|
||||
"""
|
||||
|
||||
language: Optional[Language] = None
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
*,
|
||||
api_key: str,
|
||||
base_url: str = "https://api.x.ai/v1/tts",
|
||||
voice: Optional[str] = None,
|
||||
language: Optional[str | Language] = None,
|
||||
sample_rate: Optional[int] = None,
|
||||
encoding: Optional[str] = "pcm",
|
||||
aiohttp_session: Optional[aiohttp.ClientSession] = None,
|
||||
params: Optional[InputParams] = None,
|
||||
settings: Optional[Settings] = None,
|
||||
**kwargs,
|
||||
):
|
||||
@@ -74,54 +94,25 @@ class XAITTSService(TTSService):
|
||||
Args:
|
||||
api_key: xAI API key for authentication.
|
||||
base_url: xAI TTS endpoint. Defaults to ``https://api.x.ai/v1/tts``.
|
||||
voice: Voice identifier. Defaults to ``"eve"``.
|
||||
|
||||
.. deprecated:: 0.0.105
|
||||
Use ``settings=XAITTSService.Settings(voice=...)`` instead.
|
||||
|
||||
language: BCP-47 or base language code (for example ``"en"`` or ``"pt-BR"``).
|
||||
Defaults to ``"en"``.
|
||||
|
||||
.. deprecated:: 0.0.105
|
||||
Use ``settings=XAITTSService.Settings(language=...)`` instead.
|
||||
|
||||
sample_rate: Output sample rate for PCM audio. Defaults to 24000 Hz.
|
||||
sample_rate: Audio sample rate. If None, uses default.
|
||||
encoding: Output encoding format. Defaults to "pcm".
|
||||
aiohttp_session: Optional shared aiohttp session.
|
||||
params: Deprecated input parameters object.
|
||||
settings: Runtime-updatable settings. When provided alongside deprecated
|
||||
parameters, ``settings`` values take precedence.
|
||||
settings: Runtime-updatable settings.
|
||||
**kwargs: Additional keyword arguments passed to ``TTSService``.
|
||||
"""
|
||||
default_settings = self.Settings(
|
||||
model=None,
|
||||
voice="eve",
|
||||
language="en",
|
||||
language=Language.EN,
|
||||
)
|
||||
|
||||
if voice is not None:
|
||||
self._warn_init_param_moved_to_settings("voice", "voice")
|
||||
default_settings.voice = voice
|
||||
if language is not None:
|
||||
self._warn_init_param_moved_to_settings("language", "language")
|
||||
default_settings.language = (
|
||||
self.language_to_service_language(language)
|
||||
if isinstance(language, Language)
|
||||
else language
|
||||
)
|
||||
|
||||
if params is not None:
|
||||
self._warn_init_param_moved_to_settings("params")
|
||||
if not settings and params.language is not None:
|
||||
default_settings.language = self.language_to_service_language(params.language)
|
||||
|
||||
if settings is not None:
|
||||
default_settings.apply_update(settings)
|
||||
|
||||
super().__init__(
|
||||
pause_frame_processing=True,
|
||||
sample_rate=sample_rate,
|
||||
push_start_frame=True,
|
||||
push_stop_frames=True,
|
||||
sample_rate=sample_rate or self.XAI_DEFAULT_SAMPLE_RATE,
|
||||
settings=default_settings,
|
||||
**kwargs,
|
||||
)
|
||||
@@ -130,14 +121,22 @@ class XAITTSService(TTSService):
|
||||
self._base_url = base_url
|
||||
self._session = aiohttp_session
|
||||
self._session_owner = aiohttp_session is None
|
||||
self._encoding = encoding
|
||||
|
||||
def can_generate_metrics(self) -> bool:
|
||||
"""Check if this service can generate processing metrics."""
|
||||
return True
|
||||
|
||||
def language_to_service_language(self, language: Language) -> Optional[str]:
|
||||
"""Convert a Language enum to xAI's language format."""
|
||||
return str(language)
|
||||
"""Convert a Language enum to xAI language format.
|
||||
|
||||
Args:
|
||||
language: The language to convert.
|
||||
|
||||
Returns:
|
||||
The xAI-specific language code, or None if not supported.
|
||||
"""
|
||||
return language_to_xai_language(language)
|
||||
|
||||
async def start(self, frame):
|
||||
"""Start the xAI TTS service."""
|
||||
@@ -175,7 +174,7 @@ class XAITTSService(TTSService):
|
||||
"text": text,
|
||||
"voice_id": self._settings.voice,
|
||||
"output_format": {
|
||||
"codec": self.XAI_PCM_CODEC,
|
||||
"codec": self._encoding,
|
||||
"sample_rate": self.sample_rate,
|
||||
},
|
||||
}
|
||||
@@ -189,12 +188,11 @@ class XAITTSService(TTSService):
|
||||
|
||||
measuring_ttfb = True
|
||||
try:
|
||||
async with self._session.post(self._base_url, json=payload, headers=headers) as response:
|
||||
async with self._session.post(
|
||||
self._base_url, json=payload, headers=headers
|
||||
) as response:
|
||||
if response.status != 200:
|
||||
error = await response.text(errors="ignore")
|
||||
logger.error(
|
||||
f"{self} error getting audio (status: {response.status}, error: {error})"
|
||||
)
|
||||
yield ErrorFrame(
|
||||
error=f"Error getting audio (status: {response.status}, error: {error})"
|
||||
)
|
||||
@@ -208,6 +206,11 @@ class XAITTSService(TTSService):
|
||||
if measuring_ttfb:
|
||||
await self.stop_ttfb_metrics()
|
||||
measuring_ttfb = False
|
||||
yield TTSAudioRawFrame(chunk, self.sample_rate, 1, context_id=context_id)
|
||||
yield TTSAudioRawFrame(
|
||||
chunk,
|
||||
self.sample_rate,
|
||||
1,
|
||||
context_id=context_id,
|
||||
)
|
||||
except Exception as e:
|
||||
yield ErrorFrame(error=f"Unknown error occurred: {e}")
|
||||
|
||||
Reference in New Issue
Block a user