Commit e3c5bb7f authored by Archit Tamarapu's avatar Archit Tamarapu
Browse files

[pyaudio3dtools] fix PCM support for ISM input

parent 66b1c883

scripts/pyaudio3dtools/spatialaudioconvert.py

+1 −1
Original line number Diff line number Diff line
@@ -196,7 +196,7 @@ def spatial_audio_convert( 


            # initialise metadata object for ISM

            metadata_obj = spatialmetadata.Metadata()

            metadata_obj.init_for_ism(in_file, in_fs, in_meta_files)

            metadata_obj.init_for_ism(in_file, in_fs, in_nchans, in_meta_files)



            # TODO alternative paths for binaural rendering for now

            if out_format.startswith("BINAURAL_ROOM"):

scripts/pyaudio3dtools/spatialmetadata.py

+4 −3
Original line number Diff line number Diff line
@@ -211,13 +211,13 @@ class Metadata: 
        for object_index in range(self.nb_objects):

            print(f"  Object #{object_index} Type: {self.objects[object_index]}")



    def _append_audio_array(self, audio_wav=None, fs=48000, object_index=None):

    def _append_audio_array(self, audio_wav=None, fs=48000, nchan=1, object_index=None):

        if audio_wav is None:

            audio_wav = self.audio_wav[-1]

        if object_index is None:

            object_index = -1



        x, fs = audiofile.readfile(audio_wav, fs=fs)

        x, fs = audiofile.readfile(audio_wav, fs=fs, nchannels=nchan)

        logger.debug(f"Append {audio_wav}: {x.shape[0]} by {x.shape[1]}")



        # Select appropriate channels & resample if necessary
@@ -245,6 +245,7 @@ class Metadata: 
        self,

        in_file: str,

        in_fs: int,

        in_nchan: int,

        metadata_files: list,

    ) -> None:

        self.audio_wav.append(in_file)
@@ -252,7 +253,7 @@ class Metadata: 
        for csv in metadata_files:

            self.objects.append(read_ism_ivas_data(csv, object_index=self.nb_objects))

            self.objects[-1]["track_index"] = self.nb_objects

            self._append_audio_array(self.audio_wav[-1], fs=in_fs)

            self._append_audio_array(self.audio_wav[-1], fs=in_fs, nchan=in_nchan)

            self.nb_objects += 1



    # Get audio array with sampling rate