PODCAST
Our more than 2,500 talks by nearly 100 different speakers represent a vast treasure house of first hand material by those who lived their lives wholly for God and the Master. A number of the curated talks are relatively short: from a couple of minutes to fifteen or twenty minutes (some much longer) and focused on a particular theme.

Many of those curated talks are featured in our podcast delivered daily to subscribers — click here to subscribe to the Avatar Meher Baba, Mandali Hall Talks podcast. Longer talks in our library run to well over an hour and are available only on this website, primarily in the COLLECTIONS section accessed via the main menu.

TRANSCRIBING
We have launched a new project of transcribing the 2,500+ curated talks so that a visitor to the website can follow along by reading the text of the words on their screen while listening to a given talk. We are aiming for verbatim text that identically mirrors the spoken word of the audio recordings. The idea is like making English language subtitles available on the screen as one is watching a foreign language movie. Perhaps a better analog would be watching a movie or TV show from England. We find it's often helpful to see the dialogue as a subtitle because the accents are hard to decipher. Given the incredible amount of material, we recognize this is a daunting and ambitious task. A small stipend is offered.

Nonetheless we believe there is great value in being able to read along while listening to a speaker. Studies have found that reading transcripted words while listening to a speaker vastly increases comprehension. Because the vast majority of the talks are from non-native English speakers, there is no software available at the present time that can automatically “read” a speaker’s spoken words and create anything remotely resembling intelligible English. Those involved in this project use the highest rated software, but still must play each tape in real time and listen carefully while typing out the words spoken. Many hours go into transcribing even a relatively short talk. There are volunteers enrolled in this effort right now. Even with the best software and hardware, they find that the going is slow.

To go to an example of a recently completed transcript, click here. Once you are at the page, find the title, “1. Manonash” underneath the color photo of Eruch along the left side of the page. Click on the word Manonash and a pdf transcript of that talk will open. Then back on the webpage, click the play button to hear the talk. You can then use the transcript to follow along with the recording.

Stipend policy

Due to a generous gift from a benefactor, transcribers can elect to opt in to the Stipend Program to receive a small amount in remuneration for their work. Our current policy is to offer US $1 for every uploaded page that has been transcribed. The amount is payable only in US dollars.

Please contact us for more information.

NOISE REDUCTION
Since nearly all our talks were recorded by amateurs using amature microphones and recording equipment and sometimes at quite a distance from the speaker, some talks are difficult to listen to because of technical limitations. Other talks have fairly decent sound quality but are difficult to decipher due to the perhaps unfamiliar accent of the person speaking. For those reasons and more, we are attempting to accomplish what is sometimes called “audio sweetening“ for all the talks in the repository.

For our purposes that process boils down to noise reduction which is the process of removing noise from a signal. All signal processing devices, both analog and digital, have traits that make them susceptible to noise. Noise can be random or white noise with an even frequency distribution, or frequency dependent noise introduced by a device's mechanism or signal processing algorithms. In electronic recording devices, a major type of noise is hiss created by random electron motion due to thermal agitation at all temperatures above absolute zero. Also annoying can be background chatter, noise from outside the room filtering into the Hall or even cross-talking that may be occurring while Eruch or Mani or another mandali is speaking on a point to those listening in Mandali Hall. This noise reduction process involves using sophisticated audio software to filter out unwanted sounds. The process is not exclusively automatic; it involves a lot of manual labor and several passes for each talk to adjust the settings correctly in order to get the clearest, most listenable reproduction.

CATALOGING & CURATING
An important feature of the website is named “COLLECTIONS”. There are 3 collections which contain nearly 750 full length talks; a given talk could run anywhere from ½ hour to 1½ hours in length of unedited raw material. The current collections available are the Avatar Collection, the Andy Muir Collection, and the Awakener Collection. All can be accessed under COLLECTIONS in the main navigation menu. While the material in the Collections is priceless, there is currently only partial cataloging for the Andy Muir Collection and the Awakener Collection. A more complete description of individual talks appears for the Avatar Collection. A huge amount of raw material has been transcribed and its contents are now much easier to access. What is needed is a culling through of these full length talks into smaller discrete bites focused on specific themes. Once curated, those shorter talks can be incorporated into the podcasting rotation.

While cataloging would be useful for the 2,500+ curated talks, it is also needed for these much longer talks.

For the partially cataloged Andy Muir Collection, you will see that for each audio file (talk) on the left hand side of the page, there is a description of the talk on the right, alongside it. For the Avatar Collection, the audio file’s description can be accessed by clicking the TEXT button alongside each talk of the 340 talks in the collection. When clicked, the TEXT button opens a pdf file onto your computer or device so you can get a good idea of the contents of the talk, or in the case of a talk that has been verbatim transcribed, you can read along as you listen to the talk.

While cataloging of the longer talks in the Collections is clearly needed, most of the shorter, already existing curated talks currently have only a title to inform the visitor as to their contents and would also benefit from fuller descriptions.