WO2005043914A1

WO2005043914A1 - Transcription system and method

Info

Publication number: WO2005043914A1
Application number: PCT/AU2004/001381
Authority: WO
Inventors: Nicole Gilvear; Nari Jennings
Original assignee: Caption It Pty. Ltd
Priority date: 2003-10-23
Filing date: 2004-10-12
Publication date: 2005-05-12

Abstract

A transcription system (10) for providing captions to the soundtrack of a program for broadcast, including a broadcast site (22), and a remote captioning site (12) linked by a secure communications channel in a virtual private network (16) and incorporating respective codec devices (22.4, 12.4) to enable real-time transfer of audio data and re-transmission of caption data with minimal delay to broadcast. A transcription method suited for use with the system is also described.

Description

TRANSCRIPTION SYSTEM AND METHOD

FIELD OF THE INVENTION

The invention relates to a transcription system and method. The invention has particular utility in the generation of captions for broadcasts and will be described hereunder in this context.

BACKGROUND ART

Captions represent in text form the dialogue of a TV program, TV commercial, video, DVD or other cinema screening, including other features of the soundtrack such as sound effects and music. Foreign language subtitles are a translation of the dialogue only.

Captions have certain style characteristics, such as colouring and positioning, (or placement on screen), to assist viewers to follow who is speaking. Captions are commonly displayed at the bottom of a TV or cinema screen but can be placed elsewhere within the screen, such as the top, to avoid covering important onscreen information or people's mouths. There are two main categories of captions, namely live captions and offline captions. Live captions are generally used for live, changeable or fast turnaround broadcast environments, such as news and current affairs or special events, such as for election coverage.

Offline captions are used for programs that are produced well in advance of airti e and that are unlikely to change once they've been produced, such as a drama series, or telemovie. These offline captions are embedded into the soundtrack of the program.

Traditional methods of live captioning require that all captioning equipment and personnel be located at the point of broadcast. This is due to the existing technological requirements that all captioning data must be broadcast from the same point as the sound and vision in order to properly synchronise the caption text.

Generally, broadcasts are transmitted from a central broadcasting point to a national network. From time to time, it may be necessary or desired to break into that national network broadcast (or feed) in order to insert local programming. This may include, for example, television stations in different States or Territories producing and broadcasting their own local news bulletins. While each State or Territory station produces a number of local stories of particular interest to their demographic, they may wish to file-share these stories with each other in order to gain sufficient content to fill their respective bulletins. In addition to the point of broadcast requirement, traditional captioning methods and software have also meant that it is necessary for a captioning facility to be located onsite at the station in order to integrate or link captioning software with newsroom computer systems, which contain vital programming information, and to gain access to the local edit booths where the local stories are produced. Therefore, in order to caption each station's news bulletin or other locally produced program content, captioning personnel and equipment must be located onsite in each respective State or Territory.

Developments in software and captioning products, particularly those with remote or browsing capabilities, have enabled captioning to be produced remotely.

However, these methods can be costly and do not necessarily deliver a quality product either to the remote site captioning personnel or the viewer. A common method is to employ a device that requires two telephone lines - one to receive audio, the other to send data. The quality of the audio is analogue, and neither the audio nor the data are handled in real-time leaving it subject to lengthy delays in transference.

One significant change to the captioning method was the introduction of a coder/decoder (codec) device, which includes a purpose-built modem for POTS (Plain Old Telephone System), for captioning operations. One example of this type of device is known as a Tieline® "Commander" and is manufactured by Tieline® Technology of Malaga, Western Australia, (www.tieline.com). The Tieline codec device allows for data to be sent and audio received at the same time on one standard telephone line. It also provides for the transference of the audio and data in effectively real-time with no delay, thus emulating the transference rate of audio/data send/receive as if at the point of broadcast. This improved transference overcomes any significant time delays that are experienced by other 'remote' methodologies.

While the introduction of remote techniques has resolved the point of broadcast issue at a national level, the ability to integrate the newsroom computer systems with captioning systems and access local edit booths remains unresolved to a large extent and/or subject to significant infrastructure cost issues. Current practice allows the captioning to be performed remotely from the point of broadcast. However, this remote site must usually be located at some other network site. This requires considerable infrastructure from the remote network site in order to address the integration issue and will still require personnel at each other site in order to facilitate access to local edit booths.

SUMMARY OF THE INVENTION

The present invention provides a transcription method and system which addresses the above drawbacks and/or provides enhanced functionality, particularly for remote captioning.

Preferably, the method and system includes connecting a remote transcription site to a broadcast site utilising a secure channel established in a virtual private network, provided in a packet-switched digital communications network, such as the Internet. Other preferred features of the invention will be apparent from the dependent claims and from the following description of the preferred embodiment.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described in a non-limiting manner with respect to a preferred embodiment in which:-

FIG 1 is an overview of the transcription system according to a first embodiment of the present invention; and

FIG. 2 is a flowchart showing the transcription method of the first embodiment.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

With reference to FIG 1 , there is illustrated a transcription system 10, that may be separated into two main functional areas; namely remote captioning and point of broadcast.

The system includes a number of captioning sites 12, 14 that are remotely located from a plurality of broadcast sites 22, 24, 26 transmitting a program or event. Each broadcast site 22, 24, 26 includes a server 22.2, 24.2, 26.2 and a codec device 22.4, 24.4, 26.4, such as the Tieline product. The codec device may be bidirectional and is conventionally configured to enable up to 15kHz bandwidth of high quality audio to be transported (at bit rates as low as 24,000bps) over a common telephone line (POTS) which typically has a bandwidth of only 3kHz.

Each codec, for example device 12.4, may conventionally be configured to accept analogue audio at two balanced inputs from a program or event soundtrack, which inputs are combined and amplitude limited prior to being digitised, preferably by a high quality 20-bit analogue-to-digital converter. The serial data bit stream is then fed to a 32-bit digital signal processor (DSP) that encodes the audio data using a digital audio compression algorithm, not dissimilar to that of the proprietary compression algorithm supplied by Tieline Technologies. After compression, the DSP then produces a low bit rate serial data stream that is fed to a V.34 data modem.

In the present embodiment of the invention and in contrast to the POTS application, compressed audio data is uploaded onto a secure channel of a virtual private network (VPN) 16 for transmission. The codec devices 12.4, 22.4 are configured to implement TCP/IP to allow a gateway to the VPN and provide a path to the network. This ensures that the compressed audio data is routed and delivered to a receiving application at a captioning site in the correct sequence.

The audio data is transported over the Internet 18 to be received at a remote codec device 22.4 located at captioning site 22. At this stage the received audio data is decompressed, fed to the digital-to-analogue converter, then amplified to line level and fed to a balanced output connector.

The amplified analogue audio data is then transcribed into text and the text is re- transmitted to a desired broadcast site incorporating the text data according to TCP/IP onto a VPN channel 16.

It will be appreciated by those skilled in the art that this process may be continuous and desirably occurs in real-time for the duration of the program or event requiring captions.

In conjunction with the VPN, browse-capable captioning software allows for the integration and remote access to the relevant broadcast server 22.2 to enable viewing of broadcast edit booths to coordinate with any program or event change. It should be appreciated, that this process of remote server access occurs in substantially real-time, and minimises the delay associated with transmission time.

Therefore, the transcription method 30 of the embodiment may be described as follows, with reference to the steps illustrated in FIG. 2: • Audio stream from broadcast site 22 is sent through VPN 16 in compressed form and received at remote captioning site 12 at step 31 ;

• Audio decompressed by remote codec device 12.4 and transcribed to text form via shorthand machine, qwerty keyboard or voice at step 32; • Transcribed text is converted into captions using specialist captioning software, such as that called WinCAPS, at step 33;

• Captions then outputted via a server 12.2, which is desirably part of a remote LAN (not shown in F1G.1 ), at step 34;

• Server 12.2 output then routed to the remote codec device 12.4 which sends caption data via the VPN 16 to the point of broadcast 22 at step 35 (while simultaneously receiving audio from the point of broadcast codec device 22.4 via the VPN 16);

• Point of broadcast VPN 16 receives caption data and sends it on to the point of broadcast codec device 22.4, and caption data is then output by the point of broadcast codec device 22.4 to an inserter (not shown in FIG. 1) at step 36;

• Inserter encodes and inserts the caption data into the vertical blanking line of the video signal which is then broadcast to the viewer at step 37;

• The data received at the point of broadcast codec 22.4 suitably includes a flag indicating that no further audio data is required to be captioned for testing at step 38, thereby allowing termination of the captioning session at step 39 or in the absence of an end flag, allowing return of control to step 32 (above), as required. A skilled person would also appreciate that this transcription transmission process can function on a local, national, and international basis, and may be applied to other real-time events requiring transcription or captioning, such as CART (Communication Access Real-time Translation), court or parliamentary proceedings.

It is to be understood that the above embodiments have been provided only by way of exemplification of this invention, and that further modifications and improvements thereto, as would be apparent to persons skilled in the relevant art, are deemed to fail within the broad scope and ambit of the present invention described and claimed herein.

Claims

1. A transceiver system for providing remote transcription services including: at least one broadcast site for broadcasting a soundtrack of a program or event to be transcribed; at least one transcribing site for transcribing program or event; a codec device corresponding to each at least one broadcast site and at least one transcribing site, configured for packet switched transmission; at least one virtual private network for providing a secure communications channel to link said at least one broadcast site with at least one transcribing site.

2. The transceiver system of claim 1 wherein packet switched transmission utilises the TCP/IP.

3. The transceiver system of claim 1 wherein the transcription services are provided in real-time.

4. The transceiver system of claim 1 wherein the virtual private network allows access to a broadcast server from the transcribing site.

5. A method for providing a remote transcription service, said method including the steps of: establishing a secure communications channel provided within a virtual private network between at least one broadcast site and at least one transcribing site; transmitting on secure communications channel through a first codec device a soundtrack of a program or event to be transcribed from at least one broadcast site; receiving on secure communications channel through a second codec device said soundtrack of program or event at said at least one transcribing site; transcribing soundtrack to text; re-transmitting text of soundtrack on a secure communications channel through second codec device to at least one broadcast site; and receiving on secure communications channel through first codec device text of soundtrack at said at least one broadcast site; wherein first and second codec devices are configured for packet switched transmission.

6. The method of claim 5 wherein packet switched transmission utilises TCP/IP.

7. The method of claim 5 wherein the transcription services are provided in real-time.

8. The method of claim 5 wherein the virtual private network allows access to a broadcast server from transcribing site.

9. A method of providing a transcription service, said method including the steps of: coupling a broadcast site to a remote transcription site utilising a secure communications channel provided within a digital communications network; encoding the audio signal at the broadcast site for transmission over the digital communications network; transmitting the encoded audio signal via the secure communication channel to the transcription site; decoding the encoded audio signal at the broadcast site; transcribing the decoded audio data into text data; and re-transmitting to the broadcast site a text data signal which corresponds to said audio signal.

10. The method of claim 9 wherein the coupling step involves connecting the transcription site to a broadcast site utilising a secure channel established in a virtual private network.

11. The method of claim 10 wherein the virtual private network is provided in a packet-switched digital communications network.

12. The method of claim 11 wherein the digital communications network includes a public network such as the Internet.