a
hj) @ 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 ZdZdZdZdZdZdZdZe ddZ
dddd fdd
Z d
d Z d d
Z
eeeeeef dddZeee eeeeef dddZeeee f eeeef eeeef f d fdd ZdddZdd 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).
TF )Zmax_new_tokensN)vocoder
sampling_rate no_processorc s t j|i | || _| jdkr*tdd | _| jjt v rb|d u r\t
t
| jjn|| _|| _| jd ur~| jjj| _| jd u r| jj}| jjdd }|d ur|| dD ] }t||d }|d ur|| _q| jd u r| jst| jdr| jjj| _d S )Ntfz5The TextToAudioPipeline is only available in PyTorch.generation_config)Z sample_rater feature_extractor)super__init__r Z framework
ValueErrorr model __class__r
valuesr Zfrom_pretrainedDEFAULT_VOCODER_IDtodevicer config__dict__getupdateto_dictgetattrhasattr processorr ) selfr r r argskwargsr Z
gen_configZsampling_rate_namer `/var/www/html/assistant/venv/lib/python3.9/site-packages/transformers/pipelines/text_to_audio.pyr a s0
"z TextToAudioPipeline.__init__c K sv t |tr|g}| jjjdkrH| jjddddddd}|| |}| j rT| j
n| j }||fi |dd i}|S )
NZbarkZmax_input_semantic_lengthr
FT
max_length)r+ Zadd_special_tokensZreturn_attention_maskZreturn_token_type_idspaddingZreturn_tensorspt)
isinstancestrr r Z
model_typer Zsemantic_configr r r tokenizerr$ )r% textr' Z
new_kwargs preprocessoroutputr) r) r*
preprocess s
z TextToAudioPipeline.preprocessc K s | j || jd}|d }|d }| j rn| j || jd}d|vrL| j|d< || | jjf i ||}n2t|rtd| | jf i ||d }| j
d ur|
|}|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 ) Z_ensure_tensor_on_devicer r Z can_generater r generatelenr keysr )r% Z model_inputsr' r5 r6 r3 r) r) r* _forward s&
z TextToAudioPipeline._forward) text_inputsr5 returnc K s d S Nr) r% r; r5 r) r) r* __call__ s z TextToAudioPipeline.__call__c K s d S r= r) r> r) r) r* r? s )r; r< 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 r? r> r( r) r* r? s c C sr t | dd d ur| j|d< t | dd d ur>| j|d< | j|d< |rF|ni |rP|ni d}|d u rdi }i }|||fS )Nassistant_modelassistant_tokenizerr0 )r5 r6 )r" r@ r0 rA )r% Zpreprocess_paramsr5 r6 paramsZpostprocess_paramsr) r) r* _sanitize_parameters s
z(TextToAudioPipeline._sanitize_parametersc C sj i }| j r8t|tr |d }qDt|tr2|d }qD|}n | j|}|jdtjd |d< | j
|d< |S )Nwaveformr cpu)r Zdtypeaudior ) r r. dicttupler$ decoder torchfloatnumpyr )r% rF Z output_dictrD r) r) r* postprocess s
z TextToAudioPipeline.postprocess)NNN)__name__
__module__ __qualname____doc__Z_load_processorZ_pipeline_calls_generateZ_load_image_processorZ_load_feature_extractorZ_load_tokenizerr Z_default_generation_configr r4 r: r r/ r rG r? listr rC rM
__classcell__r) r) r( r* r s2 5&! $
r )typingr r r Z
generationr utilsr baser rJ Zmodels.auto.modeling_autor
Z!models.speecht5.modeling_speecht5r r r r) r) r) r* s