Control of Loudness in Digital TV

Thomas Lund

Outline

Control of Loudness in Digital TV

Thomas Lund

2006

Abstract

To facilitate better consistency between programs and stations, ITU, EBU and ARIB have investigated the standardization of broadcast loudness. This paper examines some consequences of a global loudness standard with regard to metering and control at the Ingest, Production and Transmission stages. Findings are reported from the latest research into mono, stereo and multichannel loudness measurement of real-world broadcast sounds. The improvements achieved by the new loudness models are quantified against previous level descriptors, such as, for example, PPM and Leq(A). Besides from reducing consumer annoyance with jumping levels, less engineering time needs being spent per audio stream. This too, is important because digital broadcast means a significant proliferation of the number of channels and the number of platforms. Each platform, such as TV, radio, internet, podcast, and other personal entertainment systems, has its own requirements for dynamic range, frequency range and speec...

Figures (15)

Fig 1. Dynamic Range Tolerance for consumers under different listening situations. While DTV in itself must be able to cover several consumer situations, other emerging digital broadcast platforms widen the dynamic range target even further.

Table 1. Typical noise levels measured by the author. All environments are realistic for broadcast consumption today. In situations with significant backgrou nd noise, such as inside various transports or urban environments, see Table 1, it’s a challenge to get a wid message across - be it music or s reproduction distortion being added, listener’s ears. The latter is becomi recent studies suggest that headphone 5-10 dB above the same person’s e dynamic range poken - without or damaging the ng important as evels may be set preference when listening through speakers, see Fig 2. f the same holds true in noisy environments, where iPods are often used, mobile platforms could pose a threat to hearing.

Fig 3. Typical noise spectrum in a moving car with the windows closed (upper trace), and when idling (lower trace). Fig 3 shows spectral noise conditions inside a car [3]. Low frequency noise from the road-tire contact is the main source, as long as the windows are kept closed.

Fig 4. Weighting filters used in combination with Leq measures. A, B, C, D, M and RLB (green). The tests suggested that a relatively simple Leq measure close to a C weighing, labeled “teq(RLB)”, under certain conditions was a good predictor of perceived loudness.

Fig 5. A dose approach to Loudness. Audio segment 1, 2 and 3 may produce the same dose measure, even though their level profiles over time are quite different.

Fig 6. A guide to reading the Loudness Model Evaluation Diagram of Fig 7. Finally, it should be noted that the idea of a perceptually based level calculation is not new. An aging, but respectable measure such as “CBS Loudness”, is still being used with success for automated level control [9]. This model has served as a Je facto reference for objective loudness measurement, in the broadcast community.

Fig. 7. Evaluation of different Loudness Models (names at the bottom) using a wide range of broadcast audio material [8]. Loudness models to the left are in better agreement with human listeners than models to the right of the chart. Red indication at the top signifies outlier audio segments, misjudged by more than 6 aB of a particular loudness model.

Fig 8. Example of Loudness Meter combining a realtime measure in the outer ring with a history in the “radar view”.

Routing internally at the station is based on linear digital audio, typically using AES/EBU and/or SDI transports. Fig 9. Example of Dolby LM100 meter measurement before and after automatic loudness correction during transmission. Challenging 20 sec broadcast segments butt edited over 5:30 minutes.

¥. OlAllONS USING Metadata OMY fOr Film Normal content transmitted using automatic realtime control of loudness and format (fixed metadata). Feature films transmitted with or without dynamic range correction (dynamic metadata). Fig. 10. Three different ways of handling Loudness Control, Multichannel audio and Data Reduction in digital broadcast. DTV transmission relies solely on metadata when it comes to loudness control and speech intelligibility. In Fig 10, drawing no. 2, the Ingest Gate (i2) is used to datareduce import programming, and to inspect metadata associated with it. Downstream of Ingest, metadata must always be available and preserved, meaning no analog transfers or sample rate converters. Routing internally at the station is based exclusively on datareduced, synchronous digital audio. Data encoders and decoders are used for breakouts and monitoring. Audio/video synchronization needs special attention in designs where an arbitrary number of monitoring posts are needed. In production studios, metadata has to be attached to all programs. Production can be native mono, stereo or 5.1 as required. OB and Live production can be incorporated using fixed metadata with appropriate upstream dynamics processing.

The main part of dynamic range translation and loudness control should be done at the station, leaving only smaller corrections to be performed at the consumer. Fig 11. Example of dynamic range re-mapping: From Home Theatre/DVD to Living Room listening conditions (Fig 1).

Fig 12. Example of dynamic range re-mapping: From Home Theatre/DVD to Living Room listening conditions (Fig 1). Fig 11 and fig 12 show rational transfer characteristics complying with the DRT of the consumer, without affecting level already on target.

REFERENCES Fig 13. Example of multiband dynamic range re-mapping of a 5.1 feature film to domestic listening conditions (Fig 7). Black curve: Center channel. Orange curve: L, R, Ls, Rs.

References (13)

Brixen, E.B.: Report on Listening Level in Headphones. Document KKDK-068-01-ebb-1 for the Danish Radio, Copenhagen, 2001.
Kässer, J. & Blum, P.: Audio Reproduction in Cars. Proceedings of Tonmeistertagung 18, Karlsruhe 1994.
Katz, B.: An Integrated Approach to Metering, Monitoring and Level Practices. JAES, no. 9, 2000.
Lund, T.: Monitoring Audio for Digital Broadcast. Proceedings of NAB BEC, Las Vegas, April 2005.
Skovenborg, Quesnel & Nielsen: Loudness Assessment of Music and Speech. Proceedings of the AES 116 convention, Berlin 2004. Preprint 6143.
ITU-R, WP6P: Audio Metering Characteristics Suitable for the Use in Digital Sound Production, Geneva, 2000.
Skovenborg, Quesnel & Nielsen: Evaluation of Different Loudness Models with Music and Speech. Proceedings of the AES 117 convention, San Francisco 2004. Preprint 6234.
Jones, B.L. & Torick, E.L.: A New Loudness Indicator for Use in Broadcasting. Proceedings of the AES 71 Convention, Montreux, 1982.
Soloudre G. & Lavoie M.: Stereo and Multi-channel Loudness Perception and Metering. Proceedings of the AES 119 Convention, NYC, 2005. Preprint 6618.
Nielsen, S. & Lund, T.: Level Control in Digital Mastering. Proceedings of the AES 107 convention, New York 1999. Preprint 5019.
Nielsen, S. & Lund, T.: Overload in Signal Conversion. Proceedings of the AES 23 conference, Copenhagen, 2003.
Lund, T.: Distortion to The People. Proceedings of the Tonmeistertagung 23, Leipzig, November 2004. Paper A05.
Submission to the Australian Broadcasting Authority: Loud Advertisements on Television. Audio, Video & Post Production Industries of Australia, 2002.

Control of Loudness in Digital TV

Sign up for access to the world's latest research

Abstract

Related papers

References (13)

Related papers

Related topics