feat: Support MULAW audio encoding feat: Support MP3_64_KBPS audio encoding feat: Support timepointing via SSML <mark> tag

PiperOrigin-RevId: 323424211 Source-Author: Google APIs <noreply@google.com> Source-Date: Mon Jul 27 13:05:41 2020 -0700 Source-Repo: googleapis/googleapis Source-Sha: a94df49e8f208649f2f5cb39a84668c6a3434ce8 Source-Link: googleapis/googleapis@a94df49

Author: yoshi-automation

google/cloud/texttospeech_v1beta1/__init__.py

@@ -24,6 +24,7 @@
 from .types.cloud_tts import SynthesisInput
 from .types.cloud_tts import SynthesizeSpeechRequest
 from .types.cloud_tts import SynthesizeSpeechResponse
+from .types.cloud_tts import Timepoint
 from .types.cloud_tts import Voice
 from .types.cloud_tts import VoiceSelectionParams
 
@@ -37,6 +38,7 @@
  "SynthesisInput",
  "SynthesizeSpeechRequest",
  "SynthesizeSpeechResponse",
+ "Timepoint",
  "Voice",
  "VoiceSelectionParams",
  "TextToSpeechClient",

google/cloud/texttospeech_v1beta1/types/__init__.py

@@ -24,6 +24,7 @@
  VoiceSelectionParams,
  AudioConfig,
  SynthesizeSpeechResponse,
+ Timepoint,
 )
 
 
@@ -36,4 +37,5 @@
  "VoiceSelectionParams",
  "AudioConfig",
  "SynthesizeSpeechResponse",
+ "Timepoint",
 )

google/cloud/texttospeech_v1beta1/types/cloud_tts.py

@@ -31,6 +31,7 @@
  "VoiceSelectionParams",
  "AudioConfig",
  "SynthesizeSpeechResponse",
+ "Timepoint",
  },
 )
 
@@ -52,7 +53,9 @@ class AudioEncoding(proto.Enum):
  AUDIO_ENCODING_UNSPECIFIED = 0
  LINEAR16 = 1
  MP3 = 2
+ MP3_64_KBPS = 4
  OGG_OPUS = 3
+ MULAW = 5
 
 
 class ListVoicesRequest(proto.Message):
@@ -128,14 +131,26 @@ class SynthesizeSpeechRequest(proto.Message):
  audio_config (~.cloud_tts.AudioConfig):
  Required. The configuration of the
  synthesized audio.
+ enable_time_pointing (Sequence[~.cloud_tts.SynthesizeSpeechRequest.TimepointType]):
+ Whether and what timepoints should be
+ returned in the response.
  """
 
+ class TimepointType(proto.Enum):
+ r"""The type of timepoint information that is returned in the
+ response.
+ """
+ TIMEPOINT_TYPE_UNSPECIFIED = 0
+ SSML_MARK = 1
+
  input = proto.Field(proto.MESSAGE, number=1, message="SynthesisInput")
 
  voice = proto.Field(proto.MESSAGE, number=2, message="VoiceSelectionParams")
 
  audio_config = proto.Field(proto.MESSAGE, number=3, message="AudioConfig")
 
+ enable_time_pointing = proto.RepeatedField(proto.ENUM, number=4, enum=TimepointType)
+
 
 class SynthesisInput(proto.Message):
  r"""Contains text input to be synthesized. Either ``text`` or ``ssml``
@@ -270,9 +285,37 @@ class SynthesizeSpeechResponse(proto.Message):
  include the WAV header. Note: as with all bytes fields,
  protobuffers use a pure binary representation, whereas JSON
  representations use base64.
+ timepoints (Sequence[~.cloud_tts.Timepoint]):
+ A link between a position in the original request input and
+ a corresponding time in the output audio. It's only
+ supported via ``<mark>`` of SSML input.
+ audio_config (~.cloud_tts.AudioConfig):
+ The audio metadata of ``audio_content``.
  """
 
  audio_content = proto.Field(proto.BYTES, number=1)
 
+ timepoints = proto.RepeatedField(proto.MESSAGE, number=2, message="Timepoint")
+
+ audio_config = proto.Field(proto.MESSAGE, number=4, message=AudioConfig)
+
+
+class Timepoint(proto.Message):
+ r"""This contains a mapping between a certain point in the input
+ text and a corresponding time in the output audio.
+
+ Attributes:
+ mark_name (str):
+ Timepoint name as received from the client within ``<mark>``
+ tag.
+ time_seconds (float):
+ Time offset in seconds from the start of the
+ synthesized audio.
+ """
+
+ mark_name = proto.Field(proto.STRING, number=4)
+
+ time_seconds = proto.Field(proto.DOUBLE, number=3)
+
 
 __all__ = tuple(sorted(__protobuf__.manifest))

scripts/fixup_texttospeech_v1beta1_keywords.py

@@ -41,7 +41,7 @@ class texttospeechCallTransformer(cst.CSTTransformer):
  CTRL_PARAMS: Tuple[str] = ('retry', 'timeout', 'metadata')
  METHOD_TO_PARAMS: Dict[str, Tuple[str]] = {
  'list_voices': ('language_code', ),
- 'synthesize_speech': ('input', 'voice', 'audio_config', ),
+ 'synthesize_speech': ('input', 'voice', 'audio_config', 'enable_time_pointing', ),
 
  }
 

synth.metadata

@@ -11,8 +11,8 @@
  "git": {
  "name": "googleapis",
  "remote": "https://github.com/googleapis/googleapis.git",
- "sha": "55094be6405640329ddc93730962b9f7e68a0fc1",
- "internalRef": "314438331"
+ "sha": "a94df49e8f208649f2f5cb39a84668c6a3434ce8",
+ "internalRef": "323424211"
  }
  },
  {