o
wi( @ sv d dl mZmZmZ ddlmZ ddlmZ ddlm Z e r/d dl
Z
ddl m Z dd l
mZ d
ZG d d d e ZdS )
)AnyUnionoverload )GenerationConfig)is_torch_available )PipelineN)%MODEL_FOR_TEXT_TO_SPECTROGRAM_MAPPING)SpeechT5HifiGanzmicrosoft/speecht5_hifiganc
s e Zd ZdZdZdZeddZdddd fdd
Zd d
Z d d Z
e d
e de
dee e
f fddZe d
ee de
deee e
f fddZd
ee ee f deee e
f eee e
f f f fdd Z dddZdd Z ZS )TextToAudioPipelinea
Text-to-audio generation pipeline using any `AutoModelForTextToWaveform` or `AutoModelForTextToSpectrogram`. This
pipeline generates an audio file from an input text and optional other conditional inputs.
Unless the model you're using explicitly sets these generation parameters in its configuration files
(`generation_config.json`), the following default values will be used:
- max_new_tokens: 256
Example:
```python
>>> from transformers import pipeline
>>> pipe = pipeline(model="suno/bark-small")
>>> output = pipe("Hey it's HuggingFace on the phone!")
>>> audio = output["audio"]
>>> sampling_rate = output["sampling_rate"]
```
Learn more about the basics of using a pipeline in the [pipeline tutorial](../pipeline_tutorial)
You can specify parameters passed to the model by using [`TextToAudioPipeline.__call__.forward_params`] or
[`TextToAudioPipeline.__call__.generate_kwargs`].
Example:
```python
>>> from transformers import pipeline
>>> music_generator = pipeline(task="text-to-audio", model="facebook/musicgen-small", framework="pt")
>>> # diversify the music generation by adding randomness with a high temperature and set a maximum music length
>>> generate_kwargs = {
... "do_sample": True,
... "temperature": 0.7,
... "max_new_tokens": 35,
... }
>>> outputs = music_generator("Techno music with high melodic riffs", generate_kwargs=generate_kwargs)
```
This pipeline can currently be loaded from [`pipeline`] using the following task identifiers: `"text-to-speech"` or
`"text-to-audio"`.
See the list of available models on [huggingface.co/models](https://huggingface.co/models?filter=text-to-speech).
T )max_new_tokensN)vocoder
sampling_rate no_processorc s t j|i | || _| jdkrtdd | _| jjt v r1|d u r.t
t
| jjn|| _|| _| jd ur?| jjj| _| jd u rm| jj}| jjdd }|d ur[|| dD ]}t||d }|d url|| _q]| jd u r| jst| jdr| jjj| _d S d S d S d S )Ntfz5The TextToAudioPipeline is only available in PyTorch.generation_config) sample_rater feature_extractor)super__init__r framework
ValueErrorr model __class__r
valuesr from_pretrainedDEFAULT_VOCODER_IDtodevicer config__dict__getupdateto_dictgetattrhasattr processorr ) selfr r r argskwargsr!
gen_configsampling_rate_namer a/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/transformers/pipelines/text_to_audio.pyr \ s4
z TextToAudioPipeline.__init__c K sv t |tr|g}| jjjdkr$| jjddddddd}|| |}| j r*| j
n| j }||fi |dd i}|S )
Nbarkmax_input_semantic_lengthr
FT
max_length)r3 add_special_tokensreturn_attention_maskreturn_token_type_idspaddingreturn_tensorspt)
isinstancestrr r!
model_typer semantic_configr# r$ r tokenizerr( )r) textr+
new_kwargs preprocessoroutputr/ r/ r0
preprocess s
z TextToAudioPipeline.preprocessc K s | j || jd}|d }|d }| j r7| j || jd}d|vr&| j|d< || | jjdi ||}nt|rDtd| | jdi ||d }| j
d urZ|
|}|S )N)r forward_paramsgenerate_kwargsr zYou're using the `TextToAudioPipeline` with a forward-only model, but `generate_kwargs` is non empty. For forward-only TTA models, please use `forward_params` instead of `generate_kwargs`. For reference, the `generate_kwargs` used here are: r r/ ) _ensure_tensor_on_devicer r can_generater r$ generatelenr keysr )r) model_inputsr+ rD rE rB r/ r/ r0 _forward s&
z TextToAudioPipeline._forward text_inputsrD returnc K d S Nr/ r) rM rD r/ r/ r0 __call__ z TextToAudioPipeline.__call__c K rO rP r/ rQ r/ r/ r0 rR rS c s t j|fi |S )a
Generates speech/audio from the inputs. See the [`TextToAudioPipeline`] documentation for more information.
Args:
text_inputs (`str` or `list[str]`):
The text(s) to generate.
forward_params (`dict`, *optional*):
Parameters passed to the model generation/forward method. `forward_params` are always passed to the
underlying model.
generate_kwargs (`dict`, *optional*):
The dictionary of ad-hoc parametrization of `generate_config` to be used for the generation call. For a
complete overview of generate, check the [following
guide](https://huggingface.co/docs/transformers/en/main_classes/text_generation). `generate_kwargs` are
only passed to the underlying model if the latter is a generative model.
Return:
A `dict` or a list of `dict`: The dictionaries have two keys:
- **audio** (`np.ndarray` of shape `(nb_channels, audio_length)`) -- The generated audio waveform.
- **sampling_rate** (`int`) -- The sampling rate of the generated audio waveform.
)r rR rQ r. r/ r0 rR s c C sr t | dd d ur
| j|d< t | dd d ur | j|d< | j|d< |r#|ni |r(|ni d}|d u r2i }i }|||fS )Nassistant_modelassistant_tokenizerr> )rD rE )r& rT r> rU )r) preprocess_paramsrD rE paramspostprocess_paramsr/ r/ r0 _sanitize_parameters s
z(TextToAudioPipeline._sanitize_parametersc C sj i }| j r t|tr|d }nt|tr|d }n |}n| j|}|jdtjd |d< | j
|d< |S )Nwaveformr cpu)r dtypeaudior ) r r: dicttupler( decoder torchfloatnumpyr )r) r] output_dictrZ r/ r/ r0 postprocess s
z TextToAudioPipeline.postprocess)NNN)__name__
__module__ __qualname____doc___load_processor_pipeline_calls_generater _default_generation_configr rC rL r r; r r^ rR listr rY re
__classcell__r/ r/ r. r0 r s0 5&! (
r )typingr r r
generationr utilsr baser ra models.auto.modeling_autor
!models.speecht5.modeling_speecht5r r r r/ r/ r/ r0 s