<-Go to AI Workflows       Go to ToC       Speech Restoration System->

1     Functions 2     Reference Model 3     I/O data of AI Workflow
4     Functions of AI Modules 5     I/O Data of AI Modules 6     AIW, AIMs, and JSON Metadata

1      Functions

Preservation of audio assets recorded on analogue media is an important activity for a variety of application domains, in particular cultural heritage. Preservation goes beyond mere A/D conver­sion. For instance, the magnetic tape of an open reel may hold important information: it can carry annotations (by the composer or by the technicians) or it can include multiple splices and/or display several types of Irregularities (e.g., corruptions of the carrier, tape of different colour or chemical composition). This information shall be preserved for a correct playback. Nevertheless, some errors can occur during the digitisation as well as the digitisation could be partial because of the corruption of the carrier. These errors shall be restored to make the content listenable. The ARP Use Case (see 5.1.5) concerns the creation of a digital copy of the digitized audio of open reel mag­netic tapes for long-term preservation and of an access copy (restored, if necessary) for correct playback of the digitized recording.

In this Audio Recording Preservation Use Case, two files are fed into a preservation system:

The following is not required:

1. Alignment of the start and end times of the two files. However, the maximum tolerated misalignment is 10s.
2. Presence of signal at the start and the end of the two files.
3. Alignment of the Useful Signal on both files.
4. The same time base for both files. However, the time difference between the same samples in two files shall not be more than 30ms for a 1-hour audio tape.

The output of the restoration process is composed by:

  1. Preservation Master Files.
  2. Access Copy Files.

2      Reference Model and its operation

Figure 1 depicts the Audio Recording Preservation Reference Model.

Figure 1 – Audio Recording Preservation Reference Model

The sequence of operations of the Audio Recording Preservation unfolds as follows:

  1. The analogue audio signal from the open-reel tape recorder is digitised as Preservation Audio File.
  2. Preservation Audio-Visual File is the combination of:
    1. The video camera pointed at the playback head of the open-reel tape recorder.
    2. The analogue audio signal digitised with the same video clock.
  3. Audio Analysis for Preservation:
    1. Detects Irregularities.
    2. Assigns IDs to them that are unique to the analysed open-reel tape.
    3. Receives an Irregularity File from the Video Analysis for Recording
    4. Extracts the Audio Files corresponding to each Irregularity received or detected.
    5. Sends the Audio Files and the Irregularity File related to all Irregularities to the Tape Irregularity Classification.
  4. Video Analysis for Preservation:
    1. Detects Irregularities.
    2. Assigns IDs to them that are unique to the analysed open-reel tape.
    3. Receives an Irregularity File from the Audio Analysis for Recording and the offset between Preservation Audio File and the Preservation Audio-Visual File.
    4. Extracts the Irregularity Images corresponding to each Irregularity received or detected.
    5. Sends the Irregularity Images and the Irregularity File related to all Irregularities to the Tape Irregularity Classification.
  5. Tape Irregularity Classification:
    1. Receives an Irregularity File with the corresponding Images and Audio Files.
    2. Classifies and selects the ones considered relevant.
    3. If the Irregularity was detected by the Video Analysis for Recording, the selected Irregularity File and the corresponding Irregularity Images are sent to the Packaging for Audio Recording.
  6. The Tape Audio Restoration uses the Irregularity File to identify and restore portions of the Preservation Audio File.
  7. The Packaging for Audio Preservation collects the Preservation Audio File, Restored Audio Files, the Editing List, the Irregularity File and corresponding Irregularity Images if detected by the Video Analyser, and the Preservation Audio-Visual File and then it produces the Preservation Master Files and Access Copy Files.

3      I/O data of AI Workflow

Table 1 gives the input and output data of Audio Recording Preservation.

Table 1 – I/O data of Audio Recording Preservation

Input Comments
Preservation Audio File A Preservation Audio File obtained by digitising the analogue tape audio recording composed of music, soundscape or speech read from a magnetic tape.
Preservation Audio-Visual File A Preservation Audio-Visual File produced by a camera pointed to the playback head of the magnetic tape recorder.
Output data Comments
Preservation Master Files Set of files providing the information stored in an audio tape recording without any restoration. As soon as the original analogue recordings is no more accessible, it becomes the new item for long-term preservation.
Access Copy Files Set of Audio Files derived from the Preservation Audio File, where potential speed, equalisation or reading backwards errors that occurred in the digitisation process have been corrected.

4      Functions of AI Modules

The AIMs required by this Use Case are specified in Table 2.

Table 2 – Functions of AI Modules of Audio Recording Preservation

AIM Function
Audio Analysis for Preservation 1. At the start, it calculates the offset between Preservation Audio and the Audio of the Preservation Audio-Visual File.
2. Sends Audio Irregularity File to and receives Video Irregularity Files from Video Analysis for Preservation.
3. Extracts the Audio Files corresponding to the Irregularities identified in both Irregularity Files.
4. Sends the Irregularity merged from the Audio and Video Irregularity Files to Tape Irregularity Classification with the corresponding Audio Files.
Video Analysis for Preservation 1. Detects and enters the Video Irregularities of the Preservation Audio-Visual File in the Video Irregularity File.
2. Sends Video Irregularity File to and receives Audio Irregularity Files from Audio Analysis for Preservation.
3. Extracts the Images corresponding to the Irregularities of both Irregularity Files.
4. Sends the Irregularity merged from the Audio and Video Irregularity Files to Tape Irregularity Classification with the corresponding Video Files.
Tape Irregularity Classification 1. Receives Irregularity File (Audio) and Audio Files from Audio Analysis for Preservation.
2. Receives Irregularity File (Video) and Irregularity Images from Video Analysis for Preservation.
3. Classifies and selects the relevant Irregularities of the Preservation Audio-Visual File and Preservation Audio File.
4. Sends the Irregularity File related to the selected Irregularities to Tape Audio Restoration.
5. Sends the Irregularity Files related to the selected Irregularities and the corresponding Irregularity Images to Packaging for Audio Recording.
Tape Audio Restoration 1. Detects and corrects speed, equalisation and reading backwards errors in Preservation Audio File.
2. Sends Restored Audio Files and Editing List to Packaging for Audio Preservation
Packaging for Audio Preservation Produces Preservation Master Files and Access Copy Files.

5      I/O Data of AI Modules

Table 3 – CAE-ARP AIMs and their data

AIM Input Data Output Data
Audio Analysis for Preservation Preservation Audio File
Preservation Audio-Visual File
Irregularity File		 Audio Files
Audio Irregularity File
Video Analysis for Preservation Preservation Audio-Visual File
Audio Irregularity File		 Video Irregularity File
Irregularity Images
Tape Irregularity Classification Irregularity Audio Files
Audio Irregularity File
Irregularity Images
Video Irregularity File		 Irregularity File
Irregularity Images
Tape Audio Restoration Irregularity File
Preservation Audio File		 Editing List
Restored Audio Files
Packaging for Audio Preservation Preservation Audio File
Restored Audio Files
Editing List
Irregularity File
Irregularity Images
Preservation Audio-Visual File		 Access Copy Files
Preservation Master Files

6      AIW, AIMs, and JSON Metadata

Table 4 – Acronyms and URs of JSON Metadata

AIW AIMs Name JSON
CAE-ARP Audio Recording Preservation File
CAE-AAP Audio Analysis for Preservation File
CAE-VAP Video Analysis for Preservation File
CAE-TIC Tape Irregularity Classification File
CAE-TAR Tape Audio Restoration File
CAE-PAP Packaging for Audio Preservation File

 

<-Go to AI Workflows       Go to ToC       Speech Restoration System->