+hSSKrSSKJrJr SSKJr SSKrSrSr SS K J r "S S \\5rg!\ ar SrS\ S 3r Sr C N&Sr C ff=f) N) ConfigMixinregister_to_config)SchedulerMixinTFzCannot import librosa because zB. Make sure to correctly install librosa to be able to install it.)Imagec@\rSrSrSrSr\SS\S\S\S\S\S \S \4S jj5rS\S\4S jr SS\ S\ R4Sjjr S\4SjrSS\S\ R4SjjrS\4SjrS\S\R"4SjrS\R"S\ R4SjrSrg )Mel%a Parameters: x_res (`int`): x resolution of spectrogram (time). y_res (`int`): y resolution of spectrogram (frequency bins). sample_rate (`int`): Sample rate of audio. n_fft (`int`): Number of Fast Fourier Transforms. hop_length (`int`): Hop length (a higher number is recommended if `y_res` < 256). top_db (`int`): Loudest decibel value. n_iter (`int`): Number of iterations for Griffin-Lim Mel inversion. zmel_config.jsonx_resy_res sample_raten_fft hop_lengthtop_dbn_itercXPlX0lX@lX`lXplUR X5 SUl[(d[[5eg)N) rsrrrrset_resolutionaudio_librosa_can_be_imported ValueError _import_error)selfr r rrrrrs l/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/deprecated/audio_diffusion/mel.py__init__ Mel.__init__:sH%    E) '']+ +(cXlX lURUlURUR-S- Ulg)zSet resolution. Args: x_res (`int`): x resolution of spectrogram (time). y_res (`int`): y resolution of spectrogram (frequency bins). N)r r n_melsr slice_size)rr r s rrMel.set_resolutionPs2  jj **t6:rN audio_file raw_audioc Ub)[R"USURS9uUlnOX l[ UR5UR UR -:aj[R"UR[R"UR UR -[ UR5- 45/5Ulgg)zLoad audio. Args: audio_file (`str`): An audio file that must be on disk due to [Librosa](https://librosa.org/) limitation. raw_audio (`np.ndarray`): The raw audio file as a NumPy array. NT)monor) librosaloadrrlenr rnp concatenatezeros)rr$r%_s r load_audioMel.load_audio^s  !#LL$477KMDJ"J tzz?TZZ$//9 9RXXtzzDOO?[^abfblbl^m?m>o5p(qrDJ :rreturncF[UR5UR-$)zlGet number of slices in audio. Returns: `int`: Number of spectograms audio can be sliced into. )r*rr"rs rget_number_of_slicesMel.get_number_of_slicesps4::$//11rslicecZURURU-URUS--$)zGet slice of audio. Args: slice (`int`): Slice number of audio (out of `get_number_of_slices()`). Returns: `np.ndarray`: The audio slice as a NumPy array. r )rr")rr6s rget_audio_sliceMel.get_audio_sliceys+zz$//E1DOOuqy4QRRrcUR$)zDGet sample rate. Returns: `int`: Sample rate of audio. )rr3s rget_sample_rateMel.get_sample_rates wwrc[RRURU5URUR UR URS9n[R"U[RURS9nX0R-S-UR- RSS5S-R[R5n[R "U5nU$)zConvert slice of audio to spectrogram. Args: slice (`int`): Slice number of audio to convert (out of `get_number_of_slices()`). Returns: `PIL Image`: A grayscale image of `x_res x y_res`. )yrrrr!)refrrg?)r(featuremelspectrogramr8rrrr! power_to_dbr+maxrclipastypeuint8r fromarray)rr6Slog_Sbytedataimages raudio_slice_to_imageMel.audio_slice_to_images OO * *""5)dggTZZTXTcTclplwlw + ##A266$++Fkk)S04;;>DDQLsRZZ[][c[cd) rrLc[R"UR5SS9RURUR 45nUR S5UR-S- UR- n[R"U5n[RRRX@RURURUR S9nU$)zConverts spectrogram to audio. Args: image (`PIL Image`): An grayscale image of `x_res x y_res`. Returns: audio (`np.ndarray`): The audio as a NumPy array. rG)dtypefloatr@)rrrr)r+ frombuffertobytesreshapeheightwidthrFrr( db_to_powerrAinverse mel_to_audiorrrr)rrLrKrJrIrs rimage_to_audioMel.image_to_audios==@HH%,,X]XcXcIde(4;;6rpsb C: #MN+~N$ (+mns/AAA