Skip to content

elevenlabs

__all__ = ['ElevenLabsTextToSpeechDriver'] module-attribute

ElevenLabsTextToSpeechDriver

Bases: BaseTextToSpeechDriver

Source code in griptape/drivers/text_to_speech/elevenlabs_text_to_speech_driver.py 
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
@define
class ElevenLabsTextToSpeechDriver(BaseTextToSpeechDriver):
    api_key: str = field(kw_only=True, metadata={"serializable": True})
    voice: str = field(kw_only=True, metadata={"serializable": True})
    output_format: str = field(default="mp3_44100_128", kw_only=True, metadata={"serializable": True})
    _client: Optional[ElevenLabs] = field(default=None, kw_only=True, alias="client", metadata={"serializable": False})

    @lazy_property()
    def client(self) -> ElevenLabs:
        return import_optional_dependency("elevenlabs.client").ElevenLabs(api_key=self.api_key)

    def try_text_to_audio(self, prompts: list[str]) -> AudioArtifact:
        audio = self.client.generate(
            text=". ".join(prompts),
            voice=self.voice,
            model=self.model,
            output_format=self.output_format,
        )

        content = b""
        for chunk in audio:
            content += chunk

        # All ElevenLabs audio format strings have the following structure:
        # {format}_{sample_rate}_{bitrate}
        artifact_format = self.output_format.split("_")[0]

        return AudioArtifact(value=content, format=artifact_format)

_client = field(default=None, kw_only=True, alias='client', metadata={'serializable': False}) class-attribute instance-attribute

api_key = field(kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

output_format = field(default='mp3_44100_128', kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

voice = field(kw_only=True, metadata={'serializable': True}) class-attribute instance-attribute

client()

Source code in griptape/drivers/text_to_speech/elevenlabs_text_to_speech_driver.py 
23
24
25
@lazy_property()
def client(self) -> ElevenLabs:
    return import_optional_dependency("elevenlabs.client").ElevenLabs(api_key=self.api_key)

try_text_to_audio(prompts)

Source code in griptape/drivers/text_to_speech/elevenlabs_text_to_speech_driver.py 
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
def try_text_to_audio(self, prompts: list[str]) -> AudioArtifact:
    audio = self.client.generate(
        text=". ".join(prompts),
        voice=self.voice,
        model=self.model,
        output_format=self.output_format,
    )

    content = b""
    for chunk in audio:
        content += chunk

    # All ElevenLabs audio format strings have the following structure:
    # {format}_{sample_rate}_{bitrate}
    artifact_format = self.output_format.split("_")[0]

    return AudioArtifact(value=content, format=artifact_format)