Morphing Audio Files

1.  Morphing audio files in Altered Studio is the process of changing your input voice into a new voice.

2.  Please note, if you want to morph different sections of audio file into different voices (e.g. for different speakers in a conversation etc), then you must highlight each section separately (on the waveform or transcription) and add a new Morph effect layer for each section. Once a section has been used to add a Morph effect layer, any subsequent morphs made in that layer will only use the audio from the original selection. For example, to create a conversation with three speakers you would need to select three separate sections of the waveform/transcript and create a new morph effect layer for each voice.

3.  Before you can morph audio you need to install your desired voices from the Voice Library (refer to Using the Voice Library).

4.  Select your desired section of the waveform or transcript and use + Effect to add a new Morph effect.

5.  On the Morph Properties panel, use the (+) button to open the Voices selection screen. From here choose one or more Voices which you would like to use and click OK.

6.  Each Voice has a number of options as outlined below, take care to set these before proceeding. Refer to Working with Presets below for instructions on how to save Voice configuration settings for future use.

Voice Model Selection

7.  The Model dropdown allows you to choose between the different models available for each Voice. These will vary depending on your subscription level and you may not have access to all types.

8.  Narration voices have greater constraint over dynamics and performance characteristics, to give a more typically narrative feel.

9.  Performance voices allow a closer reproduction of the source performance and dynamics, with the additional Prosody setting for greater performance range.

10.  Flexi voices have additional settings such as Age, Gender and Loudness to provide more variety and range.

11.  Clone voices are designed to sound more like the target voice than the Performance voices do, and are less influenced by the input speaker than the other voices.

12.  Refer to the Quota FAQ for more information on how changing Voice Models will consume quota.

Using Voice Settings

13. The Morph panel contains several other settings that will change the morph output.

14. Use 48kHz generates the output of the synthesis at 48kHz, if this is not selected then the output will be generated at 24kHz by default.

15. Decreak corrects for minor vocal fry/creak in the source file, to provide a clearer synthesis output. Leave this setting off if you prefer these sounds to come across to the synthesis from the source audio.

16. Pitch is used to change the pitch of the morphed sample from the target voice’s natural pitch. The best results come from using a range of +/- 2 semitones, however the setting allows for a wider range if you wish to experiment.

17. Prosody is available for voices other than Narration, and is used to adjust the weighting between the input voice and the target voice’s natural performance characteristics. A higher Prosody will reduce the likeness to the original performance. Closer to 0% will capture more of the input voice performance. Closer to 100% will focus more on the target voice's natural performance.

18. Text Reinforcement is available on certain subscription levels for all voices other than Clone. It can be used to correct mispronunciations on the morph output by forcing the spoken words to match the transcript in most cases. Note this is not a TTS solution, it looks at the text to make very minor corrections and fix mispronunciations. If the transcript is wrong you can edit the text before you use the Text Reinforcement functionality. NB you must have a correct transcript layer in the History before you add the Morph effect in order to use Text Reinforcement.

Additional Controls for Flexi Voices (beta)

19. There are additional controls available to the Flexi Voices. These are under ongoing development and may yield inconsistent results, however when used in combination they can deliver unique voices for your project.

20. Loudness Shift changes the output between whispering (dial to the left) and loud speaking (dial to the right).

21. Age Shift changes the output between younger sounding (dial to the left) and older sounding (dial to the right).

22. Gender Shift changes the output between more feminine sounding (dial to the left) and more masculine sounding (dial to the right).

Synthesising a Morph Sample

23. Once you have finalised your settings click Synthesize to create a Morph sample. The synthesis is non-deterministic for all voice models other than Fast, so there may be slight differences between outputs generated with the same settings. For this reason we recommend generating a few samples so you can choose your preferred one.

24. Click on your chosen sample to select it, you can listen to the input audio at any time by pressing (Play Icon) on the Input Audio row in the sample bay.

25. The Mixing panel will show options for mixing, loudness and crossfade for the morphed audio with the rest of the waveform. Refer to Adding Audio Effects above for more information on using these effects.

26. Once a morph sample is selected, future edits and effects applied to this section of the waveform will be based on the morphed audio.

27. If you add a morph effect which overlaps a previous morph effect layer in the History then you will see a yellow area above the voice selection panel which says “Input for this effect already contains morphed audio”. This indicates the overlap exists, if you do not check “Use original audio” then any overlapping synthesis will result in a morph of the original morph, which may have a slight decrease in quality.

28. To avoid morphing a morph either check the “Use original audio” button to base the synthesis on the audio layer immediately before the overlapped morph layer, or delete the new layer and create it again with a non-overlapping section.

Additional Morph Sample Options

29. By right-clicking on a morph sample you can perform the following functions.

30. Open the sample as a new tab to edit it separately

31. Copy the sample to the Clipboard for future use in the same session (refer to Working with the Clipboard).

32. Export Preset will create a new preset with the settings used to create this morph sample.

33. Retake Sample will create a new sample with the same settings.

34. Export as… lets you export a wav file of the sample to disc.

35. Delete will delete the sample