With MacSpeech’s MacSpeech Scribe (http://www.macspeech.com), you can create text documents directly from spoken-word audio files. In other words, it’s a personal transcription solution for Mac OS X 10.6 (“Snow Leopard”).
It allows you to transcribe your recorded voice into written text with up to 99% accuracy. A built-in Recognition Editor helps increase accuracy by letting you correct your text and train your profile as you use the app.
MacSpeech Scribe is designed for a single person to transcribe a recording of a single person’s voice. Alas, the technology simply doesn’t exist today to transcribe interactive interviews with two people speaking to each other due to the technical limitations of analyzing multiple voices using a single voice profile.
Still, it’s a good time for speech technology products. Within the past decades computers have gained enough power to reliably convert speech into written text. Just about every modern Mac ships with enough horsepower to do this.
MacSpeech Scribe automatically creates an individual speech profile, in one of 13 different English dialects, from an initial recorded sample of the person’s voice. A single MacSpeech Scribe license can create and use up to six individual voice profiles, each compatible with all accepted audio file formats. MacSpeech Scribe accepts audio files in .wav, .aif, .aiff, .m4v, mp4, or .m4a formats.
When composing a letter or other document by voice, you can include spoken punctuation, and MacSpeech Scribe will automatically include the punctuation marks in the output text. Audio file quality directly impacts the accuracy achieved and as such MacSpeech Scribe customers are encouraged to minimize background sound and produce spoken-word audio files using a high-quality recording device such as one of the certified recording devices listed at http://www.macspeech.com/recording/ .
When you’re speaking you should use your natural voice. Some folks have a tendency to look away from their computer screen while dictating. Too much of that isn’t a good thing. Also, while a natural voice is good, MacSpeech Scribe works better if you practice proper enunciation.
MacSpeech Scribe is available immediately, in English only, at a suggested retail price of $149.
So how is it different from MacSpeech Dictate? Both are personal solutions, but think of it this way: MacSpeech Dictate is TV, and MacSpeech Scribe is TiVo. In other words, both are your voice, but one operates in real time, while the other operates time-shifted via an audio recording that your or an assistant can use.
Also, unlike Dictate, Scribe doesn’t interact with other apps. Instead, all text output from the transcription process is placed into a “Note Pad” from which you can copy and paste or save the output to a text file.