Praat is a free speech analysis programme (available to download from www.praat.org) which I have used for doing tonal and segmental analysis. It is a flexible programme with many different features and I have only used the basic ones. For tonal analysis you can choose to show pitch and set it to show an appropriate pitch range, depending on the speaker. For analysis of vowel quality you can set it to show formants. You can also set the range of the spectrogram that you want to look at (e.g. I usually have it set to 0-5000Hz but if I was looking at high frequency sounds such as silibant fricatives I would set it to show up to 10,000Hz).
I sometimes use praat to split up long sound files into shorter pieces; you can export a section of a soundfile to a new file (wav, aiff, aifc, next/sun, nist or flac). One drawback is that it cannot read or write to
mp3 files.
I have found praat particularly useful for adding text to a sound file. You can open a text file in the same window as the sound file with up to three text tiers underneath the spectrogram. You can then break each text tier into sections which allows you to play just one section of the sound file and transcribe/label underneath it. It recognises all IPA characters. This is intended for phonetic analysis, however I personally find this function
very useful for transcribing texts; I usually split the tiers into sections about one sentence long and use one tier for the transcription and one for an English free translation, then I later copy these into FLEX. You need to
be careful when using a text file because it does not remind you to save it when you close the file, therefore if you are not careful you can easily loose work.
