def phasevocoder(channels,
speed=1.,
frame_length=2048,
analysis_hop=None,
synthesis_hop=None,
phase_locking=PhaseLocking.IDENTITY):
"""Returns a :class:`~audiotsm.base.tsm.TSM` object implementing the phase
vocoder time-scale modification procedure.
In most cases, you should not need to set the ``frame_length``, the
``analysis_hop`` or the ``synthesis_hop``. If you want to fine tune these
parameters, you can check the documentation of the
:class:`~audiotsm.base.analysis_synthesis.AnalysisSynthesisTSM` class to
see what they represent.
:param channels: the number of channels of the input signal.
:type channels: int
:param speed: the speed ratio by which the speed of the signal will be
multiplied (for example, if ``speed`` is set to 0.5, the output signal
will be half as fast as the input signal).
:type speed: float, optional
:param frame_length: the length of the frames.
:type frame_length: int, optional
:param analysis_hop: the number of samples between two consecutive analysis
frames (``speed * synthesis_hop`` by default). If ``analysis_hop`` is
set, the ``speed`` parameter will be ignored.
:type analysis_hop: int, optional
:param synthesis_hop: the number of samples between two consecutive
synthesis frames (``frame_length // 4`` by default).
:type synthesis_hop: int, optional
:param phase_locking: a phase locking strategy.
:type phase_locking: :class:`PhaseLocking`, optional
:returns: a :class:`audiotsm.base.tsm.TSM` object
"""
# pylint: disable=too-many-arguments
if synthesis_hop is None:
synthesis_hop = frame_length // 4
if analysis_hop is None:
analysis_hop = int(synthesis_hop * speed)
analysis_window = hanning(frame_length)
synthesis_window = hanning(frame_length)
if phase_locking == PhaseLocking.NONE:
# No phase locking: all freqyency bins are considered to be peaks
peak_finder = all_peaks
elif phase_locking == PhaseLocking.IDENTITY:
peak_finder = find_peaks
else:
raise ValueError(
'Invalid phase_locking value: "{}"'.format(phase_locking))
converter = PhaseVocoderConverter(channels, frame_length, analysis_hop,
synthesis_hop, peak_finder)
return AnalysisSynthesisTSM(converter, channels, frame_length,
analysis_hop, synthesis_hop, analysis_window,
synthesis_window)
def wsola(channels,
speed=1.,
frame_length=1024,
analysis_hop=None,
synthesis_hop=None,
tolerance=None):
"""Returns a :class:`~audiotsm.base.tsm.TSM` object implementing the WSOLA
(Waveform Similarity-based Overlap-Add) time-scale modification procedure.
In most cases, you should not need to set the ``frame_length``, the
``analysis_hop``, the ``synthesis_hop``, or the ``tolerance``. If you want
to fine tune these parameters, you can check the documentation of the
:class:`~audiotsm.base.analysis_synthesis.AnalysisSynthesisTSM` class to
see what the first three represent.
WSOLA works in the same way as OLA, with the exception that it allows
slight shift (at most ``tolerance``) of the position of the analysis
frames.
:param channels: the number of channels of the input signal.
:type channels: int
:param speed: the speed ratio by which the speed of the signal will be
multiplied (for example, if ``speed`` is set to 0.5, the output signal
will be half as fast as the input signal).
:type speed: float, optional
:param frame_length: the length of the frames.
:type frame_length: int, optional
:param analysis_hop: the number of samples between two consecutive analysis
frames (``speed * synthesis_hop`` by default). If ``analysis_hop`` is
set, the ``speed`` parameter will be ignored.
:type analysis_hop: int, optional
:param synthesis_hop: the number of samples between two consecutive
synthesis frames (``frame_length // 2`` by default).
:type synthesis_hop: int, optional
:param tolerance: the maximum number of samples that the analysis frame can
be shifted.
:type tolerance: int
:returns: a :class:`audiotsm.base.tsm.TSM` object
"""
# pylint: disable=too-many-arguments
if synthesis_hop is None:
synthesis_hop = frame_length // 2
if analysis_hop is None:
analysis_hop = int(synthesis_hop * speed)
if tolerance is None:
tolerance = frame_length // 2
analysis_window = None
synthesis_window = hanning(frame_length)
converter = WSOLAConverter(channels, frame_length, synthesis_hop,
tolerance)
return AnalysisSynthesisTSM(converter, channels, frame_length,
analysis_hop, synthesis_hop, analysis_window,
synthesis_window, tolerance,
tolerance + synthesis_hop)
def ola(channels, speed=1., frame_length=256, analysis_hop=None,
synthesis_hop=None):
"""Returns a :class:`~audiotsm.base.tsm.TSM` object implementing the OLA
(Overlap-Add) time-scale modification procedure.
In most cases, you should not need to set the ``frame_length``, the
``analysis_hop`` or the ``synthesis_hop``. If you want to fine tune these
parameters, you can check the documentation of the
:class:`~audiotsm.base.analysis_synthesis.AnalysisSynthesisTSM` class to
see what they represent.
:param channels: the number of channels of the input signal.
:type channels: int
:param speed: the speed ratio by which the speed of the signal will be
multiplied (for example, if ``speed`` is set to 0.5, the output signal
will be half as fast as the input signal).
:type speed: float, optional
:param frame_length: the length of the frames.
:type frame_length: int, optional
:param analysis_hop: the number of samples between two consecutive analysis
frames (``speed * synthesis_hop`` by default). If ``analysis_hop`` is
set, the ``speed`` parameter will be ignored.
:type analysis_hop: int, optional
:param synthesis_hop: the number of samples between two consecutive
synthesis frames (``frame_length // 2`` by default).
:type synthesis_hop: int, optional
:returns: a :class:`audiotsm.base.tsm.TSM` object
"""
if synthesis_hop is None:
synthesis_hop = frame_length // 2
if analysis_hop is None:
analysis_hop = int(synthesis_hop * speed)
analysis_window = None
synthesis_window = hanning(frame_length)
converter = OLAConverter()
return AnalysisSynthesisTSM(
converter, channels, frame_length, analysis_hop, synthesis_hop,
analysis_window, synthesis_window)
def test_hanning(length, window):
"""Run tests for hanning."""
assert_almost_equal(windows.hanning(length), np.array(window))