Tutorial for Wavepad
- 1. Wavepad: Audio Editor Dr. Feda Negesse Language Technology Group Linguistic Capacity Building Training Addis Ababa University Bishoftu Dreamland Hotel May 21 -23, 2018
- 2. Objectives • To identify effective tools (software packages) for editing audio data. • To build skills of using software for editing audio data. • To edit audio data using an appropriate software. • To appreciate the application of software for editing 2
- 3. Software • Audacity • Adobe Audition • PRAAT • WAVEPAD • Others 3
- 4. Wavepad: Introduction • A sound editor program for Windows and Mac OS X. • Designed to be very easy and intuitive to use. • Easy to open or record a file and edit it. • You need time to explore powerful tools designed with the professional sound engineer in mind. • WavePad Basic Edition is free, but WavePad Master is licensed. 4
- 5. Features • Supports a number of file formats including wav (multiple codecs), mp3, real audio and many more. • A wide range of editing capabilities including Cut, Copy, Paste, Delete, Insert, Silence, AutoTrim and others. • Effects including Amplify, Normalize, Equalizer, Envelope, Reverb, Echo, Noise Reduction, Sample Rate Conversion and many more. 5
- 6. Functions • Ability to work with multiple files at the same time. • Includes a CD ripper to load audio direct from a CD-ROM. • Player includes Scrub/Cue control for precise editing. • Recorder supports pause, retake, auto trim and voice activated recording. • Supports MDI (Multiple Document Interface), which allows displaying multiple files all on one screen. 6
- 7. Workspace 7
- 8. Windows Upper window Displays the waveform of the audio file always in its full length. Lower window • Shows the waveform of the audio based on the zoom level. • To move the lower window side to side, use the horizontal scroll bar just under the lower window. 8
- 9. Basics - Working With Files Create a New Audio File • When creating a new file WavePad will prompt you to select the sample rate and channels. Open an Existing Audio File • To open and existing file press Ctrl+O, browse to locate the file you want to open and click • You can also load audio tracks from the CDs by select File -> Load Audio CD Track(s). 9
- 10. Basics… Save an Audio File • To save the current audio file with the current name press Ctrl+S. • To save the file with a different name or in a different format use the menu File -> Save As (press Ctrl+Shift+S). • Never save the file in a compressed format until you have finished all editing. • Every time you save and reload in a compressed format you lose some quality. 10
- 11. Copying files Copy file(s) to CD • Selecting this option will let you burn any open files to a CD • Just select "Copy file(s) to CD" then select the CD • Type (either Data CD or Audio CD), then you will be prompted . Send an Audio File • WavePad allows you to send the file by either email or direct internet connection. • Select how to send a file by using the menu File->Send. • WavePad will send the file as an attachment to an email. • Enter the email address of the recipient in the Email Address box. 11
- 12. Basics - Recording Recording • To start recording, press the F5 key, or the Record button. • The recording will be made straight into your currently open file, at the position of your cursor. • If you have some audio already selected then this will be replaced by your new recording. • If you don't have a file open, then recording will start in a new file. 12
- 13. Basics - Bookmarks Bookmarks • Positions within files that you might frequently want to return to. • To add a bookmark push Ctrl+B and enter the name of the position. • When you want to return to the bookmark push Ctrl+Shift+B. • To open the bookmark list, select it from the list and select Go to Bookmark in the right click menu. • Linked to an actual file, they persist even after you close WavePad or the Project. • You should delete bookmarks when they are not needed. 13
- 14. Basics - Regions Regions • Regions store a selected part of the recording. • To add a region select it, click Ctrl+R and give it a name. • Then at any point open the regions list (Ctrl+Shift+R). • Use the right click menu to play, select, rename, delete, copy to new, save the region or assemble the regions . • You can also export the selected region(s) into one folder by selecting "Export Region(s) to folder" 14
- 15. Basics - Regions • Regions can be extremely useful when working with a long recording like an interview that needs to be re-arranged. • As you find each interesting grab, select it and make it a region. • When done, use the assemble tool to put all the regions together. • Regions are not persistent. • If you close WavePad and you want to keep your region, you must save a WavePad project . 15
- 16. Editing - Convert Sample Rate Convert Sample Rate • WavePad can convert a file's sample rate or number of channels. • This change applies to the whole file (you cannot change just parts of the file). • To convert the sample rate of the current file use the menu Edit - > Convert Sample Rate. • Then select the new sample rate. • The sample rate must be between 6000 and 192000 samples per second. 16
- 17. Editing - Convert sampling rate • To convert the sampling rate the current file select Edit -> Convert Sample rate. • A typical reason for converting the sample rate down is when you know it is going to be used on the telephone (8000 hertz, mono) • You are about to compress the file for internet transmission (11025 or 8000 hertz , mono). 17
- 18. General Audio Concepts • This is a general introduction to key audio concepts for those who have not worked with sound before. • What are the important features of sound? • Were you a natural science student? • Did you like physics ? • Did you particularly like acoustics ? 18
- 19. Sound • The starting point for everything WavePad does is sound. • Vibrating air traveling very fast like a wave. • Created by a vibrating object and can be detected by an ear or a microphone. • A microphone converts these vibrations into alternating electronic voltage. • Then the computer's sound card can turn the voltage into the data used by WavePad. 19
- 20. Frequency • Number cycles completed per second. • The human ear’s capacity: 20Hz up to 20,000Hz . • In reality, most of us only hear upto about 15,000Hz. • But audio enthusiasts often claim they can hear sounds up to the 20,000Hz. • The frequency of a person's voice can be up to 300Hz. 20
- 21. Loudness, Volume, Amplitude, Level and Gain • Loudness, volume, amplitude and level mean roughly the same thing. • The more volume a sound is given the more power has been used to create it and the louder it sounds. • When adjusting the volume level of a sound , the "Gain" value signifies the amount of increase or decrease in the level. • This value can be represented in percent or in a scale called the "decibel" or "dB" scale. 21
- 22. Decibel • The human ear can hear a remarkably 0 dB to 120 decibels. • The ear perceives differences in a logarithmic way. • Sound engineers use the decibel scale (dB). • Reducing the volume level of a sound by 6dB means you are dropping the amplitude by 1/2 or the power by 1/4. • Conversely, a 6dB increase in the level corresponds to doubling the amplitude. • A 20dB drop means 1/10 of the amplitude (or 1/100 of the power). • The smallest unit of loudness change a person will notice is around + / - 3dB. 22
- 23. Decibel levels of some common sounds Sound Source Sound Pressure Level (dB) threshold of excellent youthful hearing 0 normal breathing, threshold of good hearing 10 soft whisper 30 mosquito buzzing 40 average townhouse, rainfall 50 ordinary conversation 60 busy street 70 power mower, car horn, ff orchestra 100 air hammer at 1m, threshold of pain 120 rock concert 130 jet engine at 30m 150 rocket engine at 30m 180 23
- 25. Audio Recording with Computers • The audio signal from the microphone is converted by your sound card into a series of numbers in quick succession. • You can think of these numbers as representing the pressure on the surface of the microphone at different points over time. • This process of converting audio into a series a numbers is called "sampling". • The sample rate is the number of times that the amplitude is converted to a number per second. • At CD quality recording, your computer stores 44100 numbers per second each representing the amplitude at the specific 25
- 26. Digital recording • The maximum possible frequency that can be carried in a sampled sound is exactly half of the sample rate. • In reality it is a little less. So for example, a recording made with a 44100 sampling rate will carry frequencies up to 20000Hz. • 6000 - Very low quality voice • 8000 - Telephone quality voice • 11025 - Reasonable quality voice - e.g., dictation • 22050 - Good quality voice, Reasonable quality music - e.g., multimedia CD. • 44100 - CD Quality. 26
- 27. Resolution • You might have seen terms like "8 bits" or "16 bits" when looking at sound files but are not sure what they mean. • The number of bits, like in the sample rate, is an indicator of resolution of the sound inside the file. • The more bits the better resolution. • WavePad uses 32 bits internally for optimal audio quality. • However, 16 bits is usually more than adequate for 27
- 28. Audio File Compression • High quality audio increases audio files. • Use what is known as "compression" to reduce the size of your files. • Compression reduces quality but saves space. • Do not save in a compressed form until it is really needed. • Save audio in an uncompressed form like 44100 Hz, 16 bit PCM format Wave to do further work on it. 28
- 29. Editing - Split Split File At Cursor • Use this option if you want to quickly split the current file into two small tracks. • To do this, click on the position where you want to split and select Edit->Split into two at this point. • Note that each of your new files shares the undo history of the original file, so the split can be undone in any of the resultant split files. 29
- 30. Split … Split File At Silences • Use this option if you want WavePad to auto split the file by detecting silence below a specified threshold level for a specified duration. • You can specify the threshold below which is considered silence (default -36dB) and the minimum duration which will be considered silence (default: 0ms). 30
- 31. Split … Split File At Bookmarks • This will split the file at each bookmark, creating a new file for each region between the bookmarks. Split File into its Components Channels • For example, when used a stereo file, this will result in two mono files, one for the original left channel, and one for the original right channel. • To join the channels again, see Join Mono Files to Make Stereo. • See also Convert Sample Rate and Channels and General Audio Concepts. Split File into Equal Duration • This will split the file into specified pieces with equal duration. 31
- 32. Split … Split File into Intervals • This will split the file into specified intervals. Split File Automatically • Given an interval or a number of output files, this option will find the best split points in the file. • The input will only approximately determine the length or the number of the output files. 32
- 33. Editing - Trim • This is useful when you have just recorded a file but there is silence or noise before the start or after the end. • Select the part of the file you want to keep and then press Ctrl+T. Trim Start (Ctrl+M) • To delete everything before the current position select Edit -> Trim -> Trim Start. 33
- 34. Trim Trim End (Ctrl+E) • To delete everything after the current position select Edit -> Trim -> Trim End. Auto Trim Silence from Start and End • Auto Trim removes the silence at the beginning and the end of the selected region. 34
- 35. Effects - Amplify Amplify • To 'amplify' is to increase the loudness or volume of the selected region. • To make a part of the recording softer or louder, select it and then use the menu Effects -> Amplify. • The volume is entered in percent (100 being no change, 50 being -6dB softer or 200 being +6dB louder). 35
- 36. Effects - Normalize Normalize • To 'normalize' is to adjust the volume so that the loudest peak is equal to the maximum signal that can be used in digital audio. • Usually you normalize files to 100% as the last stage in production to make it the loudest possible without distortion. • Another reason to normalize is to have multiple tracks sound equally loud, or to have equal average loudness. 36
- 37. Audio Cleanup - Noise Reduction Spectral Subtraction • Slow but accurate • Usually used where noise is really a problem Multiband Noise Gates • Fast and less accurate • Usually just automatically on batch voice recording jobs. • Both methods require a powerful machine. 37
- 38. Noise Reduction Spectral Subtraction • Automatic Method • This approach leaves WavePad to estimate what is noise and what is not. • It usually works well on voice and is nice and easy to use just select the region and apply the effect. • Manual Method • 1. Select a short part of 'noise only'. Usually this is from a gap in the audio. • 2. Select Effects -> Noise Reduction -> "Grab Noise Sample From Selected Area". • 3. Select the entire file. • 4. Select Effects -> Noise Reduction -> "Apply Spectral Subtraction Based 38
- 39. Noise Reduction Multiband Noise Gates • To use the all you need to do is enter the level below which you expect noise. • Usually this is between -30dB and -20dB. • If not enough noise is reduced increase the value. • If too much is reduced decrease it. 39
- 40. Noise Reduction Noise Gate • A noise gate is a filter which controls the volume of an audio signal. • Any part of your audio which is below the Threshold will be attenuated by the amount you specify. Threshold • Audio falling below this threshold will be attenuated. Hold • The period of time (in milliseconds) to wait before applying the attenuation. Release • The period of time (in milliseconds) taken to fully apply the attenuation. Attack • The period of time (in milliseconds) taken to fully remove the attenuation. Attenuation • The amount to attenuate the audio signal when it falls below the threshold. 40
- 41. Audio Cleanup - High-Pass Filter • A high-pass filter (sometimes called a low cut filter) removes all low frequencies below a specified Hz. • This is useful if you want to make your recording sound 'clearer' or less 'muddy'. • It is very usual to use a high-pass filter of about 300Hz on all voice recordings to improve intelligibility. 41
- 42. Audio Cleanup - Low-Pass Filter • A low-pass filter removes all high frequencies above a specified Hz. • This is useful if you want to make your recording sound 'clearer’. • It is very usual to use a low-pass filter of about 1600Hz on all voice recordings to improve intelligibility 42
- 43. Audio Cleanup - Automatic Gain Control • Normal recordings can have the volume of the recording too high in parts and too soft in parts. • 'Automatic Gain Control' reduces the too loud parts and increases the too soft parts. • This is sometimes a better alternative to normalization. • To use AGC select all (Ctrl+A) then use the menu Effects -> Automatic Gain Control. 43
- 44. Audio Cleanup - DC Offset Correction • Often when you record audio using bad electronics, the recording has a constant 'DC' level throughout the file. • Because the ear cannot hear this you will not notice it until you attempt to edit in other audio when you can hear horrible clicks. • If you think this is the problem you can run DC Offset Correction over the entire recording before you begin to edit. • Another way to deal with this problem is to run a high pass filter (say at 50Hz) over the recording. 44
- 45. End of the discussion • Questions 45