Efficient audio description systems and methods

A method for generating an audio description of a media file using a computer system comprising a display and memory coupled to the display, the method comprising:

• storing, by the computer system, a media file comprising video data accessible via a time index and audio data synchronized with the video data via the time index;
• storing a transcript of the audio data comprising transcription data synchronized with the video data via the time index;
• rendering, via the display, one or more images from portions of the video data;
• rendering, via the display, text from portions of the transcription data in synchrony with the one or more images;
• rendering, via the display, at least one cell within the text, the at least one cell being associated with at least one point within the time index;
• receiving input identifying that least one point within the time index;
• receiving input specifying audio description text to associate with the at least one point;
• synthesizing the audio description text to generate audio description data comprising audio data having at least one renderable duration;
• storing the audio description data;
• and
• storing an association between the audio description data and the at least one point.

A non-transitory computer readable medium storing computer-executable sequences of instructions to generate an audio description of a media file via a computer system, the sequences of instructions comprising instructions to:

• store, in a memory, a media file comprising video data accessible via a time index and audio data synchronized with the video data via the time index;
• store, in the memory, a transcript of the audio data comprising transcription data synchronized with the video data via the time index;
• render, via a display, one or more images from portions of the video data;
• render, via the display, text from portions of the transcription data in synchrony with the one or more images;
• render, via the display, at least one cell within the text, the at least one cell being associated with at least one point within the time index;
• receive input identifying the at least one point within the time index;
• receive input specifying audio description text to associate with the at least one point;
• synthesize audio description text to generate audio description data comprising audio data having at least one renderable duration;
• store, in the memory, the audio description data;
• and
• store, in the memory, an association between the audio description data and the at least one point.

A transcription system configured to generate audio description snippets and a snippet manifest from a source media file, the transcription system comprising:

• a time-coded transcript of the source media file;
• and
• a synthesized audio video interface configured to
• display the source media file,
• display the time-coded transcript,
• receive input identifying a selected time location within the source media file,
• receive input specifying audio description text to associate with the selected time location, the audio description text having at least one text characteristic;
• generate an estimated duration of the audio description text using at least one of the at least one text characteristics;
• display the estimated duration of the audio description text;
• generate an audio snippet from the audio description text;
• store the audio snippet as a file;
• and
• store, in the snippet manifest, the audio snippet, the audio description text, the selected time location in the source media file, and a duration of the snippet.