2019-09-13: The text limits become tighter
At the end of January I announced French Phonetic Dictionary Project. As a quick reminder, the goal is to create a free open-source dictionary that would contain the phonetic transcription of French words in the following format:
Several volunteers offered their help. Almost immediately we realised that the database contains a lot of missing transcriptions and a lot of very rare words. Since we are not very numerous, it would be unreasonable to waste our time trying to find the phonetic transcription, let's say, of the French word "alkékenge" (which is a plant by the way). So we decided to limit our efforts first to the most frequent French words (10,000).
I also wanted to use the offered help more wisely, so I decided to create an automatic tool that would convert French words into their phonetic transcription according to the general pronunciation rules in French. Since I am a bit busy now, in China, it took me almost two months to create this tool.
In addition to general rules for each letter (for example, "é" is pronounced as [e] and "ç" as [s]), I also described more than 160 exceptions to these rules. I continue to add more exceptions in order to improve the quality of the conversion. Right now the tool gives the correct transcription of French words in 90-94% of cases. So basically what we (volunteers) are doing right now is just comparing the discrepancies between the phonetic transcription in the database and the transcription obtained with the automatic tool. If the database doesn't contain the phonetic transcription for any particular word, the volunteer needs to double-check whether the automatic tool gave the correct transcription.
Today I am releasing the first version of the French Phonetic dictionary. It contains 3,432 words in the following intervals: abaissement-attribution, cargaison-éventail, grisâtre-irrésistible. As I mentioned in the previous post, the dictionary is published under a Creative Commons Attribution-Share Alike 4.0 License.
Thanks to everybody who is helping with the dictionary!
Please do not give the direct link to the dictionary (the text file) – the hotlinking will not work on EasyPronunciation.com. If you want to share it with your friends, please give the link to this page.
If you want to participate, please contact me. Even if you don't want to participate, it would be very appreciated if you could share the above page with your friends, for example, via the social networks. It would help promoting the project in the search engines and attract more volunteers. This way we could finish the project sooner and at the same time improve the work of the French phonetic translator.
Speaking of the translator, there's some good news and some bad news. The good news is that the automatic tool is now integrated in the translator. That means that if a word cannot be converted using the old algorithm, it will be converted with this tool. The phonetic transcription obtained in such a way will be displayed surrounded by two signs "_". The bad news is that a part of the database is now changed and that affected the conversion of the verbs of the 3rd group (and probably other words too). So now the translator is somewhat in a transitional period. As soon as the final version of the phonetic dictionary will be released, I will be able to correct these errors.